Computational Science and Its Applications – ICCSA 2022 Workshops

Several methods and operational tools for assessing the sustainability and corresponding aspects can be identified in the current literature. At international level, the use of synthetic indices is clearly established through analytical indicators capable of expressing multiple aspects from an economic, social and environmental perspective. By a literature review, the construction of indices through a multi-criteria approach can be placed in the weights assignment and in construction processes based on the geometric and arithmetic average of values. The allocation of appropriate weights to performance indicators lacks, in particular, an objective methodology and subjective elements linked, e.g., to the decision-makers involved and corresponding interests. This research aims to describe a methodological frame for indices constructing through the multi-criteria approach of the Choquet Integral. The use of Choquet’s integral supports the evaluations of multiple aspects of sustainability as monitoring of the relative unbalanced values, and the weights assignment occurs through analytical functions well-established, as the Shapley function.

Since the current need to renovate the existing residential asset, especially in Italy numerous fiscal measures have been promoted in order to encourage the buildings energy improvement initiatives. Among the incentive policies for the energy requalification, the Italian “Relaunch” Law Decree No. 34/2020 has introduced a fiscal deduction equal to 110% (so-called “Superbonus”) that constitutes a relevant financing measure able to support property owners in activating energy retrofit operations. With reference to the issue outlined, the present research aims to develop a methodology for the economic benefits evaluation, by considering the convenience of the subjects involved associated to the energy upgrade interventions using the Superbonus mechanism. The methodology provides for five phases and it is applied with reference to the residential sector and to the three main urban areas into which the Italian city of Bari is divided (central, semicentral and peripheral).For each urban area, the market value differential between the post-energy intervention situation and ante-energy intervention one is determined, by considering the case “with” the Superbonus incentive and that “without” it. The methodology proposed could represent a valid tool to support the private operators in the determination of the urban areas or, in a more general terms, the cities for which a higher convenience in energy efficiency investments is detected.

During the last decades, the public investments have very often determined “white elephants” whose initial costs have increased during construction phase and, at the end of project realization, the necessary financial resources for the operation of the investment have not been available. Starting from these bankruptcy initiatives, different technique for an accurate ex ante planning of the investment costs, and for a detailed analysis of the financial sustainability have been implemented. The Cost-Benefit Analysis is the most used tool for the public investments validation as it is able to verify the financial intervention convenience, and it allows an optimal allocation of available resources in order to guarantee the highest return on investment in the reference area. In this research, the Cost-Benefit Analysis potential and limits have been highlighted through a specific case study.

Special Economic Zones (SEZs) are geographical areas with regulatory regime where the enterprises receive incentives, through i.e. tax breaks, and have administrative simplifications, with the aim to increase the enterprises competitiveness, the attraction of direct investments especially by foreign subjects, the new increase in exports, the creation of jobs work and, more generally, the strengthening of the productive fabric by stimulating industrial growth and innovation.A SEZ implies the implementation of regulatory and administrative simplification actions in order to rationalize the procedures and to make the relationship among administration, citizen and business simpler, transparent and direct.Calabria Region has approved Strategic Development Plan for SEZ Calabria, that has its center in Gioia Tauro including the port and the industrial area. Gioia Tauro port is specialized in container transhipment operations and it has a great expansion capacity to become a third-generation port also due to the presence of the SEZ.The paper proposes a general model, a graph, that defines the concept of administrative simplification and introduces a cost function associated with it. The paper presents a general framework of the administrative simplification system developed by Calabria Region for the SEZ. The deepening of this topic is very interesting in relation to its impacts on the economic and social sector of the region and the country.

Special Economic Zones (SEZ) are a place-based measure to boost the economic growth adopted in several countries all-over the world. SEZs are geographic areas inside a country where tax incentives, favourable land-use policies, employment subsidies, and advantageous financial assistance encourage the settlement of domestic and foreign-invested manufacturing and services for export.The paper concerns the ex-ante estimation of aggregate economic impact of a SEZ in Calabria region (Italy). SEZ in Calabria was conceived to operate in several existing industrial areas of the region, having its fulcrum around the transhipment hub port of Gioia Tauro, in order to propagate its benefits to the whole region in terms of export and employment. The aggregate economic impacts of the SEZ were quantified through two characteristic variables: exports and employment of industrial firms settled in Calabria. Two development scenarios are defined: a Do-Nothing scenario, in the absence of the SEZ and with the current economic policies operating in Calabria; and a SEZ scenario, assuming a full activation of SEZ (with complete availability of financial resources and decrees for simplification) at the end of 2018.The comparison of the two scenarios shows relevant direct positive impacts in the SEZ case, both the in terms of export and employment, in the industrial sectors that benefits of the SEZ incentives.

The attractiveness of regions and places for investments, tourists, students, talented people, and other economic actors is a relevant issue for regional economic development, due to the increasing importance of flows of firms, resources and people on the global scale. Attractiveness for peripheral regions is even harder, due to their nature of marginal and peripheral locations which causes clear locational disadvantages compared to central locations. Calabria is a case of Italian peripheral region which should raise its attractiveness to increase its level of development, improving several critical location factors, like transport accessibility, which in this region is extremely scarce.The Special Economic Zone (SEZ) is an opportunity for the development of the territory. The territorial system around the Special Economic Zone must be planned by the strategic urban plan. Only the urban plan determines the integrated quality system between infrastructures, urbanization, territory and landscape. The strategic urban plan for territories with Special Economic Zones converts urban spaces into places of development, maximizes the offer of services and technological innovation and increases the quality of life and well-being of citizens.

Special Economic Zones (SEZs) play a relevant role for countries to attract foreign investment and increase economic development. A Special Economic Zone (SEZ) can influence production efficiency through various channels. Different factors can contribute to increase the production of value-added other than the one linked with economic activities.This paper focuses on the possible contribution of research and training for increasing value-added in a SEZ context. Research Centres, and Universities, have to interchange with all productive sectors operating in the Special Economic Zones (SEZs). Systematic connections between research and production are needed. For this reason, in a strategic perspective, a set of actions can be planned and implemented with the aim to activate and consolidate interchanges of knowledge and skills among public administrations and private companies.The paper studies the role of research and training to realize a SEZ in the Calabria region (Italy). The specific SEZ constitutes an opportunity for the economic development of the existing industrial areas of the region, and particularly the transhipment hub of Gioia Tauro. The Special Economic Zone (SEZ) can support and be a booster for the potentialities of the hub port, in relation with the research centres and universities operating in the area of influence of the SEZ.

Environmental sustainability and energy efficiency are some of the most challenging objectives to be pursued in port areas. In this context, the SUSPORT project aims to provide its contribution, affecting the Adriatic area. In the initial phase, before applying new technologies/solutions to enhance port sustainability, the baseline status shall be assessed in order to evaluate the impact of tested measures. To this end, a review of the peculiarities of the main ports of the Adriatic Sea (Italian and Croatian), including the evaluation of their carbon footprint should be carried out. The present work reports the results of this phase, focusing on the main statistics of the involved ports and their greenhouse gases inventory at an aggregated level.

Quantifying the emissions produced along different supply chains is an extremely difficult challenge. However, carbon-based emissions generated by the transport sector have an extremely significant impact on environmental sustainability.To address these issues, we propose a method for estimating the carbon footprint as an indicator of the environmental sustainability of processes as it represents the total emissions produced within a given process.Therefore, the first problem we come across is developing a tool for calculating the logistic carbon footprint, which clearly defines the boundaries of application of the model and its scope of application. For this reason, the proposed tool will follow a standardized and uniform approach in order to streamline the calculation processes and make it even more efficient: the sources of emissions related to a supply chain are innumerable, so depending on the different approaches to calculating emissions, they can lead to extremely different results. In order to streamline the calculation process, the main sources of primary emissions and indirect emissions due to the supply of fuel oil have been used in a preponderant manner, and then an adjustment factor that takes into account all factors omitted from the model has been introduced. In this way, the calculation of the carbon footprint has been made uniform, homogeneous, and comparable as well as quantitatively reliable.Once the method of elaboration of the carbon footprint is framed, we will proceed to use this synthetic indicator for the analysis of environmental sustainability for different logistic processes that consider the Port of Trieste as an intermodal exchange hub for intra-Mediterranean traffic and with destination the main markets of continental Europe, taking into account different modes of transport: road, rail, and maritime transport.

In the maritime world scenario, various challenges are affecting Mediterranean container ports, which are trying to keep high their efficiency and their competitiveness through infrastructural and managerial improvements. The identification of the priority actions requires the analysis of the productivity of each port in relation to the use of its resources. This study applies Data Envelopment Analysis (DEA) and Principal Component Analysis (PCA) in order to investigate the potential factors that can affect the efficiency of Mediterranean container ports. These methods use six input variables (yard area, berth depth, number of quay cranes, equipment, berth length and distance of the port from the Suez-Gibraltar axis) and one output variable (port throughput expressed in TEUs). The results can help to highlight the potential factors of success for Mediterranean container ports and to identify future policies and management strategies aimed towards the strengthening of the analyzed context.

Small-scale LNG logistics chains have become more important for delivering LNG via shipping from large supply terminals to customers via satellite terminals. An ideal application of small-scale LNG logistics chains is in the Mediterranean basin, where the maximum distance between two ports is always less than two thousand miles. Focusing on a Tyrrhenian application case, this study develops a modeling tool capable of defining the optimal configuration for a small-scale LNG distribution network serving a set of Tyrrhenian ports organized as a cluster. The aim is to minimize total network costs, including both port entry costs and travel costs. The problem is modelled as a Vehicle Routing Problem with Draft Limits and Heterogeneous Fleet (VRPDLHF). Different network configurations are being tested to explore the transportation cost savings that could result from systemic and integrated management of LNG supply if ports were organized in a cluster. Computational results show that, by acting as an organized cluster, LNG port depots can potentially leverage their increased bargaining power during negotiations to seek reasonable import prices that can benefit from reduced transportation costs and guaranteed volume of LNG to purchase.

The paper tackles the issue related to the sea-rail intermodal transport in gateway ports. In recent years there has been a renewed interest in land connections and therefore hinge functions, recalling the logic for which the most competitive ports are those equipped with an articulated and efficient network of internal connections. This attention requires an in-depth study of port intermodality and mutual interactions with the territories served by the port. This paper is focused on rail intermodality for container handling, recognizing the gateway port as a strategic node whose competitiveness increasingly depends on sustainable intermodal transport solutions to and from the hinterland. This is also due to the growing attention to green transport solutions such as rail. After highlighting the distinctive features of gateway ports and their key functions, attention is focused on the main Italian ports that play this role. In particular, this study explores the ports of La Spezia and Trieste due to their aptitude for intermodal rail traffic in relation to the hinterland. In both cases, the share of combined sea-rail transport and the importance of rail transport use in terms of environmental sustainability will be evaluated through an ad hoc tool (EcoTransIT World).

The issue of Special Economic Zones (SEZs) has become increasingly important for underdeveloped European regions. On a world level, the SEZs have allowed the significant development of the territories concerned. The main experiences of SEZs are in the areas of ports. The SEZs became the future of the ports in the underdeveloped regions. It is possible to model some main developments that the SEZ can achieve on the basis of international results. A SEZ has a core node in the area relating to an intercontinental port and can have other nodes located in national ports. A reconstruction of the supply model is then carried out considering all the modalities that allow to connect the different internal nodes to the region and the external macro-nodes. A system of models is proposed to estimate the main impacts of a SEZ. A test case is proposed referring to the TEN-T core node of Gioia Tauro and to the proposed SEZ in the connected areas. It is then modeled the maximum potential increase in employment due to the establishment of the Calabria SEZ, subject to land constraints.

About 70% of the Earth's surface is covered by water. A “blue planet” whose essential driving force is represented by the ports that ensure the connection also of the most peripheral and insular areas. The activities of ports are configured as an element of economic development and creation of new employment. In Europe, according to the last report of 2020, maritime transport was fundamental to import and export 74% of goods, to generate employment for 1.5 million workers and to move about 420 million passengers.At the same time, port activities are responsible for several negative externalities that are often not considered in business strategies. In recent decades the port industry has undergone a profound transformation because of different technical, commercial, and legal aspects. The smart port concept is based on the ability of new technologies to transform port services into interactive and dynamic businesses, more efficient and transparent. Its objective is to satisfy the needs of customers and users without forgetting its responsibility to the city and the citizens.The digitization combined with careful planning and management of port operations allows us to find smart solutions to optimize logistics and environmental impacts, to promote efficiency and to enhance the safety of both seaports and coastal regions involved.This paper, after a review of the literature on smart port studies, aims to build a matrix of indicators of the sustainability of port ecosystems which will allow the measuring and ranking of ports in its Smart Port category.

The paper tackles the issue related to the sea-rail intermodal transport Important but often conflictual relations have for long times related cities and their ports. In particular, the mutual evolutions of cities and ports drew very often the attention on waterfronts and on the older port areas and facilities that are abandoned or converted to urban uses and functions. There is little attention, on the contrary, towards the new areas becoming important for ports, as the inner harbors, that host increasingly important operations that are vital for linking ports and their hinterlands in order to free quays for freight handling on the seaside. These operations include, among others, warehousing, modal shift, block-train assembly. Similarly, cities are living a complex relation among their centres, the focus of policies and attention by policymakers, and their where, on the contrary there is a general lack of services and communities suffering for deprivation and little access. These areas, furthermore, become more and more the ideal location for transport and port-related activities and facilities, among them, ideally, disused commercial and industrial sites. Such transport and logistic-related activities, however, not always are inserted into urban and transport plans, but also very often being carried on by privately-owned companies. The paper is focused on the definition of a method for highlighting the inner relations between peripheries, semi-peripheries and inner harbors, proposing a methodological analytical framework for spatially-locating the spatial pressures and potential for development. The research started from the local cases of Cagliari, Catania and Trieste as starting points for the observation of these phenomena, to extract analytical and research points for further evolutions on national and international cases.

The tourism crisis following the Covid-19 pandemic has caused many communities to rethink and review tourism. In fact, in European countries, many destinations are now focused on more inclusive and sustainable measures rather than over tourism, so they have invested in sustainability to create tourist-friendly places. The rural areas, in particular, are affected by a demand motivated by a longing for discovery and authenticity, and they seem to be working towards a multi-scalar planning: walkability and digitalization stand out as fundamental choices to meet the needs of tourists and residents. In this context, the paper aims, after a preliminary review of the literature on rural areas’ walkability, to highlight the relationship between rural walkability and tourism through the analysis and comparison of two rural paths: the Mining Path of Santa Barbara (Sardinia, Italy) and the Path of Saint James (Gran Canaria, Spain). It represents virtuous examples of fruition and enhancement of the insular landscape, promoting a place-based approach for a sustainable and cohesive local development.

AR games such as Pokemon Go, Jurassic World Alive and other games encourage us to venture out in the real open world to also see the beauty of it. However, tourism has come almost to a virtual stop during the first year of the COVID-19 pandemic and recovery is slow. We present a tile-based experience-sharing PixoMap, which incorporates some aspects of smart tourism. For user requirements gathering, we first compare factors that make popular games such as Pokemon Go, Minecraft, and the Sims popular. Findings indicate that people enjoy collecting objects, such as monsters or cards, freedom and creativity, escape and sometimes nostalgia. Our PixoMap game allows players to virtually browse an area in the map, and choose a tile. Each tile contains memories (Memors)/experiences/stories. Users can read others’ experiences and share their own experiences, play a minigame to earn in-game currency, to change his/her 2D avatar or change the tile’s color or optionally, to own the tile. Alpha user feedback confirms and refines our design. Heuristic evaluation and user experience feedback at the end of the study, are positive and encouraging.

This paper presents an iris recognition framework to recognize irises from distantly acquired face images using image gradient-based feature extraction and K-Nearest Neighbor with various distance classifiers. The work herein applies the gradient local auto-correlation descriptor to extract discriminative features from the iris images and to reduce feature dimensionality by optimizing some parameters. Several distance metrics are applied in the iris classification stage to reduce computational complexity and build the classification models. The proposed framework effectively handles the noisy artefacts, rotation, occlusion, and illumination variation challenges. The experiments are carried out on a publicly accessible CASIA-V4 distance database to ascertain the effectiveness of distant iris recognition and to compare the efficacy of several existing distant classifiers. The experimental results justify that distance metrics influence the recognition outcomes of the classifier significantly, and the recognition performance of the Correlation distance metric is better than the other distance classifiers for iris gradient features.

E-payment is an important concept of the core e-commerce elements; it has become one of the essential success factors for financial and business services. This work elaborates the roles of perceived trust and perceived security on continuance usage of e-payment through collecting data from students who have used e-payment for e-commerce in Vietnam. The maximum likelihood of structural equation modelling is analyzed on a total sample of convenient of 210 customers. Research findings demonstrate that perceived trust and perceived security have significant roles in the model of e-payment adoption. The model amounts to 26% of continuance usage of e-payment.

In recent years, machine learning techniques have been utilized in sensitive areas such as health, medical diagnosis, facial recognition, cybersecurity, etc. With this exponential growth comes potential large-scale ethical, safety, and social ramifications. With this enhanced ubiquity and sensitivity, concerns about ethics, trust, transparency, and accountability inevitably arise. Given the threat of sophisticated cyberattacks, it’s critical to establish cybersecurity trustworthy concepts and to develop methodologies and concepts for a wide range of explainable machine cybersecurity models that will assure reliable threat identification and detection, more research is needed. This survey examines a variety of explainable machine learning techniques that can be used to implement a reliable cybersecurity infrastructure in the cybersecurity domain. The main aim of this study is to execute an in-depth review and identification of existing explainable machine learning algorithms for cyberattack detection. This study employed the seven-step survey model to determine the research domain, implement search queries, and compile all retrieved articles from digital databases. This research looks at the literature on trustworthy machine learning algorithms for detecting cyberattacks. An extensive search of electronic databases such as ArXiv, Semantic Scholar, IEEE Xplore, Wiley Library, Scopus, Google Scholar, ACM, and Springer was carried out to find relevant literature in the subject domain. From 2016 to 2022, this study looked at white papers, conference papers, and journals. Only 25 research papers were chosen for this research paper describing trustworthy cybersecurity and explainable AI cybersecurity after we retrieved 800 articles from web databases. The study reveals that the decision tree technique outperforms other state-of-the-art machine learning models in terms of transparency and interpretability. Finally, this research suggests that incorporating explainable into machine learning cybersecurity models will help uncover the root causes of defensive failures, making it easier for cybersecurity experts to enhance both cybersecurity infrastructures and development, rather than just model results, policy, and management.

This paper presents the results of a comparative study of machine learning techniques when predicting deep vein thrombosis. We used the Ri-Schedule dataset with Electronic Health Records of suspected thrombotic patients for training and validation. A total of 1653 samples and 59 predictors were included in this study.We have compared 20 standard machine learning algorithms and identified the best-performing ones: Random Forest, XGBoost, GradientBoosting and HistGradientBoosting classifiers. After hyper-parameter optimization, the best overall accuracy of 0.91 was shown by GradientBoosting classifier using only 15 of the original variables.We have also tuned the algorithms for maximum sensitivity. The best specificity was offered by Random Forests. At maximum sensitivity of 1.0 and specificity of 0.41, the Random Forest model was able to identify 23% additional negative cases over the screening practice in use today.These results suggest that machine learning could offer practical value in real-life implementations if combined with traditional methods for ruling out deep vein thrombosis.

The purpose of this work is to identify the suspicious behavior of the clients of a financial institution. Financial institutions use rule-based systems to detect unusual transactions. These systems focus on individual transactions or simple transaction patterns. Due to this, the need arises to detect suspicious behavior using machine learning since many of the machine learning algorithms are designed to capture complex patterns. Descriptive analysis and predictive analysis are used to detect suspicious behavior. Within the descriptive analysis, outliers are sought in transactional movements. For the predictive analysis, we start from a set of alerts with the label of whether or not they were reported by the Compliance Unit. As a result of the descriptive model, a set of customers that have behaved in an unusual way is obtained and as a result of the predictive model, the alerts that should be reported are predicted. It is concluded that the techniques: descriptive analysis, descriptive analysis and the rule-based system can complement each other, since they focus on different aspects of the identification of unusual transactions and should not be considered as exclusive alternatives.

Software systems have become an integral part of all the organizations. These systems are performing many critical operations. A defect in these systems affects the product quality and the software development process. Prediction of the impact category of these defects helps in improving defect management process as well as taking correct decisions to fix defects. Orthogonal defect classification is a popular model for classifying defects and it provides an in-depth analysis of the defects. In this study, we proposed an auto classify approach to classify the defects into impact categories as defined by Orthogonal Defect Classification (ODC). Bag of words, term frequency-inverse document frequency and word embedding have been used to represent the textual data into numeric vectors. For experimental work, we have used 4,096 reports form three NoSQL databases. We have trained and tested the proposed autoclassify approach using Support Vector Machine (SVM) and Random Forest Classifier (RFC). We achieved maximum accuracy 94% and 85.99% using SVM and RFC respectively.

In 2020, the number of Brazil inhabitants was approximately 212 millions, whereas the number of vehicles jumped from 104 to 107 millions. It is observed, therefore, that there is 1 vehicle for every 2 inhabitants. That same year, the population in urban areas went to 80%. Managing traffic in big cities is becoming a huge challenge. Traffic lights operating with Fixed Time Signal to control vehicle flux are no longer efficient in all traffic scenarios. Technological advances in Computer Vision, moving objects detection and classification techniques and the demanding of little computer power to manage these tasks allowed the development of TEOP, a CV-based traffic control system. This low cost solution was implemented to advance the Fixed Time Signal system, cameras and logical network infrastructure already in use in Brazilian cities, transforming regular traffic lights into smart traffic lights (STL). An application was developed and installed in a computer to capture images of traffic, count vehicles and calculate time needed for them to pass through. Raspberry controlled traffic lights. In comparison to regular traffic lights, STLs improved traffic flow by 33%, allowing a wait of just 4 s in cases where there is no vehicle flow at the competing traffic light and reducing the crossing time of 46 vehicles from 152 to 116 s when there were 12 vehicles on the competing side, a significant gain. It was also capable of reporting traffic jams and creating a database that could be used for decision-taking by agencies responsible for each route.

Robotics is used successfully in medicine. Mechanical gadgets assist the most intricate procedures, providing precision, error correction, stereotactic positioning, subtle controlling, and in-place visualization, among other benefits. However, few applications are designed to perform continuously embedded within the human body. Implants such as the peacemaker that works permanently inside the patient involve high risk. However, the risk is acceptable given the benefit that peacemakers provide when working correctly. The current proposal envisages a similar scenario. One in ten babies in America develops hydrocephalus. The abnormality is more incident in premature babies, but term babies are also subject to malady. Even adults develop a sort of hydrocephalus that keeps intracranial pressure at normal levels. Hydrocephalus quickly escalates from mild to severe due to cerebrospinal fluid circuit issues (CSF) issues. The only mechanism to alleviate the pressure within the CSF containers is to release the fluid progressively. Such as purpose requires opening the skull and locating a discharging device – diverting shunt – that goes through the brain mass and reaches the ventricles. This invasive gadget grants a way out to liquid but is not always infallible. The shunt often fails, and reiteratively opening the skull is the method to pursue. Shunted patients with more the 100 brain surgeries have motivated the campaign no-more-brain-surgeries promoted by the Hydrocephalus Association ( https://www.hydroassoc.org/join-us-as-we-say-nomorebs/ ).This document registers a revision on immersive robotics and sensors that allow us to create a submergible device capable of exploring 3D printed macro models of the CSF containers. This first approach will serve as a testing land to understand fluid mechanics, navigability, and blockage removal techniques from inside a model mimicking the brain in healthy and abnormal conditions. Once this is accomplished, we intend to miniaturize the device and check the feasibility of a medical application that controls cerebrospinal fluid abnormalities while avoiding recurrent surgery.

Information is one of the most important assets for organizations. In the current scenario of a stagnant economy and fierce competition, it is imperative for organizations to be proactive, and for this to be possible, it is necessary to have timely access to the necessary information, so that it supports the decision-making process, thus allowing anticipate business strategies to react in a timely manner. It is in this context that Business Intelligence (BI) and its tools arise. Through BI it is possible to have access to an integrated set of solutions that allow data analysis in order to capture the information and knowledge necessary to leverage the business. BI allows organizations to use tools to analyze, plan, predict, solve problems, understand, innovate and learn in order to increase organizational knowledge, allowing effective actions that help to establish and achieve concrete business objectives in order to increase its efficiency and competitiveness. There is a wide variety of BI tools available on the market, which facilitate the use of the right data and visualization in ways that allow you to understand what it means. It is pertinent that organizations know how to select the appropriate tools for their needs. In this sense, a comparative analysis will be carried out between the different BI tools. The characteristics, specifications, and attributes of each one of them will be detailed to know key aspects of the tools that help the different organizations, regardless of the service they offer, to achieve their processes in an effective and efficient way. This comparison will allow companies to have a new perspective when using BI tools and generating business strategies.

This research work investigates the deployment of data sampling and ensemble techniques in alleviating the class imbalance problem in software defect prediction (SDP). Specifically, the effect of data sampling techniques on the performance of ensemble methods is investigated. The experiments were conducted using software defect datasets from the NASA software archives. Five data sampling methods (over-sampling techniques (SMOTE, ADASYN, and ROS), and undersampling techniques (RUS and NearMiss) were combined with bagging and boosting ensemble methods based on Naïve Bayes (NB) and Decision Tree (DT) classifier. Predictive performances of developed models were assessed based on the area under the curve (AUC), and Matthew’s correlation coefficient (MCC) values. From the experimental findings, it was observed that the implementation of data sampling methods further enhanced the predictive performances of the experimented ensemble methods. Specifically, BoostedDT on the ROS-balanced datasets recorded the highest average AUC (0.995), and MCC (0.918) values respectively. Aside NearMiss method, which worked best with the Bagging ensemble method, other studied data sampling methods worked well with the Boosting ensemble technique. Also, some of the developed models particularly BoostedDT showed better prediction performance over existing SDP models. As a result, combining data sampling techniques with ensemble methods may not only improve SDP model prediction performance but also provide a plausible solution to the latent class imbalance issue in SDP processes.

In recent times, data science has seen a rapid increase in the need for individuals and teams to analyze and manipulate data at scale for various scientific and commercial purposes. Groups often collaboratively analyze datasets, thereby leading to a proliferation of dataset versions at each stage of iterative exploration and analysis. Thus, an efficient collaborative system compatible with handling various versions is needed rather than the current most often used ad-hoc versioning mechanism. In a collaborative database, all the collaborators working together on a project need to interact together to perform extensive curation activities. In a typical scenario, when an update is made by one of the collaborators, it should become visible to the whole team for possible comments and modifications, which in turn aid the data custodian in making a better decision. Relational databases provide efficient data management and querying. However, it lacks various features to support efficient collaboration. In these databases, the approval and authorization of updates are based completely on the identity of the user, e.g., via SQL GRANT and REVOKE commands. In this paper, we present a framework well suited for collaboration and implemented on top of relational databases that will enable the team to manage as well as query the dataset versions efficiently.

Software sentiment analysis has applications in numerous software engineering tasks ranging from code suggestions to evaluating app reviews which help to save the development team valuable time and increase productivity. In recent years, sentiment analysis has been used to study the emotional state of developers through sources like commit messages. State-of-the-art sentiment analysis techniques have been employed to accomplish these tasks with varying results. The goal of this paper is to provide a comparison between the performance of various models for possible applications of sentiment analysis in software engineering. We have used three different datasets to account for the possible applications: JIRA, AppReviews, and StackOverflow. In this work, six word embedding techniques have been applied on above datasets to represent the text as n-dimensional vectors. To handle the skewed distribution of classes present in the data, we have employed two class balancing techniques in the form of SMOTE and Borderline-SMOTE. The resulting data is subjected to six feature selection techniques, and finally, the sentiment of the text is classified using 14 different classifiers. The experimental results suggest that some models are very successful in accurately classifying the sentiment of the text, whereas choosing the wrong combination of ML techniques can lead to disappointing performance.

Public-private partnership (PPP) is a system implemented to improve the economic value of infrastructure outputs for the citizen of a country. Knowledge management (KM) has been seen as a way to develop successful and sustainable PPPs through constant learning and continuously improving the implementation of PPP processes. However, the presence of KM can hardly be seen on PPP project websites operated in Nigeria. The study developed a web-based knowledge management framework for PPP projects in Nigeria. The study examined existing PPP websites in Nigeria and noted that they do not have knowledge management frameworks. The system design was achieved using a use case diagram, system block diagram, and activity diagram. The web-based system is designed for the general public to access PPP project details and documents to aid accountability and transparency and help plan future PPP projects in Nigeria. The ICRC Admin, contractor’s firm, and consultants on PPP projects are to populate the PPP data on the web-based knowledge management platform. The web-based knowledge management platform has six (6) primary interfaces. The essential interface is the knowledge management interface that archives the PPP project documents and is made available to users of the platform.

The use of freely available, open-source code to reduce the time needed to create new software or add functionality to existing software is a common practice. With analysis of recent high-profile cases of open-source software packages being corrupted by the original developer, or the introduction of remote back-door functionality by malicious actors, it has been shown that there is much that can be done to help with simplifying the decision-making process of using any open-source code. This paper provides the basis for a simple-to-use checklist that can be used to quickly analyze open-source libraries for its suitability within an individual’s or organization’s code base. Fourteen projects were selected at random from a popular code hosting site that made use of specific biometric security libraries. The conclusions derived from the use of the checklist and the analysis of the selected projects will help with simplifying the decision-making process of using open-source code for software projects.

Lean, as a business approach, has gained popularity in several functional areas. One of these applications is Lean Robotics that focus on the utilization of Lean aspects to improve robotic deployment. This study aims to be the first to conduct a Multivocal review on what Lean Robotics is, its main components, its benefits, and challenges and how it evolved. It was found that Lean Robotics is defined differently by some sources, and that its components can be understood both theoretically and practically. The benefits of Lean Robotics are found to resonate from the prioritization of human and machine collaboration, and the use of various Lean tools via continuous improvement. However, some challenges of Lean Robotics like cost and fear might arise if organizations are uneducated in what Lean Robotics offers regarding its knowledge.

Web interface testing is a sort of system testing level and it is laborious if accomplished manually, since it is necessary to map each of the elements that make up the interface with its respective code. Furthermore, this mapping makes test scripts very sensitive to any changes to the interface’s source code. Approaches for automated web testing have been proposed but the use of hyper-heuristics, higher-level search techniques aiming to address the generalization issues of metaheuristics, for web testing are scarce in the literature. In this article we present a multi-objective web testing method, MWTest, which automates the generation of test cases based only on the URL of the web application and a new proposed generation hyper-heuristic, called GECOMBI. The GECOMBI hyper-heuristic takes into account combinatorial designs to generate low-level heuristics to support our goal. Moreover, the implementation of the MWTest method creates a Selenium test script quickly and without human interaction, exclusively based on the URL in order to support the automated execution of test cases too. In our evaluation, we compared GECOMBI to another generation hyper-heuristic, GEMOITO, and four metaheuristics (NSGA-II, IBEA, MOMBI, NSGA-III). Results show superior performance of GECOMBI compared to the other approaches.

The data created by web users while navigating on a website constitutes graph data. Large-scale graph data is generated on websites many users visit with high frequency. Analyzing large-scale graph data using artificial intelligence techniques and predicting user behavior by creating models is an actively studied research topic. Within the scope of this research, a machine learning business process is proposed that will allow the interpretation of graph data obtained from web user navigation data. A prototype application was developed to demonstrate the usability of the proposed business process. The developed prototype application was run on graph data obtained from websites with intense user-system interaction. A comprehensive evaluation study was carried out on the prototype application. The results obtained from the empirical evaluation are promising and show that the proposed business process is used.

Conversational agents are gaining popularity in the corporate world as a way to increase customer experience and economic competitiveness. Additionally, developments in augmented intelligence systems employ natural language processing to provide the industry with natural and clear interaction experiences. Multilingual conversational bot or Chabot is important in every area of life, especially in a multicultural community recognized for its numerous accents and slang among many social groupings Moreover, most present Chabot systems only handle one language at a time, and the training session is cumbersome since it needs various dialects for different purposes. This research presents a multilingual chatbot that allows clients to converse in many languages as if they were conversing with a real person to achieve a smart enterprise management software. The proposed system was implemented using React.JS and python programming language on a Pentium III processor speed of 600 MHz minimum. The proposed multilingual service will deal with the limitation of the existing system by developing a system that will allow multiple languages on chatbot agents. The system will allow the users to converse in their languages, which will make communication easy between the system and the users. The proposed system will include a user-friendly interface that will assist in guiding each user on how to utilize it effectively without any specialized training.

Microservices are becoming increasingly popular for the development of distributed systems. However, in order to adopt microservices in practice, one has to deal especially with decomposition, communication and deployment of services. These issues can be solved by applying design patterns to microservices. Although it seems simple, developers may present difficulties when selecting the correct pattern to solve a given problem. This situation happens both to beginners and experienced professionals, as there are a considerable number of patterns to master. This study proposes an information retrieval-based approach to recommend patterns applied to microservices considering problems expressed by developers in natural language. To evaluate the proposed approach, we created two corpus, the first one containing 10 patterns, and the second 10 problems. When performing the experiments, we were able to achieve a precision rate of 60%. From this study, it is possible to support developers in selecting patterns applied to microservices, which facilitates their adoption and the development of higher quality systems, also benefiting organizations, as well as assisting researchers in future studies, presenting a possible way to recommend these patterns.

Social media has dramatically influenced and changed the rate and the nature of crime in our society. The perpetrators cut across different age groups, social standing, and beliefs. The ability to be anonymous on social media and the lack of adequate resources to fight cybercrime are catalysts for the rise in criminal activities, especially in South Africa. We proposed a system that will analyse and detect crime in social media posts or messages. The new system can detect attacks and drug-related crime messages, hate speech, and offensive messages. Natural language processing algorithms were used for text tokenisation, stemming, and lemmatisation. Machine learning models such as support vector machines and random forest classifiers were used to classify texts. Using the support vector machine to detect crime in texts, we achieved 86% accuracy and using the random forest for crime analysis, 72% accuracy was achieved.

Water shortage across the globe causes changes in the life of the human being, not to mention that water is a resource that must be preserved for future generations. The companies that distribute and charge fees for residential water services have problems performing the periodical manual readings. Thus, showcasing inaccuracy in the costumer statements at the end of every month. The following paper describes the development process of a low-cost intelligent control and monitoring system of residential water consumption. The system design uses a flow sensor that measures the influx of liquid as it passes through the pipes. Based on the value that the conventional meter marks at the beginning of the day, it is possible to determine the daily consumption. The processing of acquired data is performed with a low-cost controller (SBC). Also, the respective control actions were transmitted towards a solenoid valve, which controls the passage of water to the house. Next, an access and communication point is established applying bidirectional MQTT (Message Queue Telemetry Transport) protocol to send and receive data wirelessly through Internet of Things (IoT). Data was stored and managed on a local server. The prototype displays information through an LCD screen and in a web page. To achieve this, the server sends information such as date-time, username, meter number, etc. The results showcase that the measurements were performed as expected. Thus, validating the possibility of using it in a larger sample.

Clean Architecture (CA) aims to address the need for more loosely coupled components and better cohesion. CA focuses on preparing software engineers to write more stable, durable, and flexible applications capable of distinguishing between details (e.g., what framework it uses) and the business logic requirements. A literature review shows that considerable effort has been devoted to cataloging and solving code smells related to code, often called code smells. However, the same does not apply to architecture smells – its software architecture counterpart. Similar research regarding other programming languages such as Java, PHP, or C# represents noteworthy works in the area, but they do not address Python applications directly. This work directs efforts towards redesigning and adapting existing Python programs to the CA principles by detecting the code smells that break the CA constraints through the developed CALint tool. Moreover, this approach proposes two extended refactoring techniques to solve these smells efficiently by grouping and comparing static code analysis and reuse them to enforce Clean Architecture’s Dependency Rule programmatically. To demonstrate the feasibility of the two refactoring techniques described in this work and the CALint tool, we applied them to three different case studies. The major findings of this work include two extended refactoring techniques and the development of a tool to verify non-conformities related to the Clean Architecture dependency rule. The results show common cases where the dependency rule was violated and highlighted by the CALint tool, which are fixed with the support of refactoring steps.

Vitamins and minerals are essential micro-nutrients required by our bodies to optimize health and maintain well-being. Currently, our nutrient recommendations are one-size-fits-all or one solution for a large group of persons, however, due to the difference of each individual’s genetic background, lifestyle, and health condition, individual’s nutrition requirement is different from each other, therefore, we should compute and provide personalized nutrition supplementation solution for each person. At the beginning of this paper, we describe the importance of personalized nutrition for keeping healthy; then we give a description of our personalized nutrition intelligent service platform by analyzing genetic data, lifestyle data and physical examination data together to generate genetic interpretation report and personalized nutrition report, and further place order to produce customized nutrition packs for each customer. To promote management efficiency and reduce errors in business processes, we developed multiple business process management systems such as laboratory information management system (LIMS), bioinformatic pipelines, genetic interpretation system, customer relationship management (CRM) system and etc. All the systems were integrated together to have the ability of processing tens of thousands of samples in parallel.

Power grids continue to develop and it is increasingly difficult to guarantee the quality of service offered to the user. In several developing countries, consumption is calculated on the basis of visual inspection, which is prone to errors. Consequently, this document outlines the construction of electrical consumption telemetering equipment. This is designed to reduce human error through manual measures and have a web backup that can be accessed from anywhere. To develop the prototype voltage and current sensors are used, and the signal is conditioned for the control stage. The processing unit is the Arduino Mega embedded board, which incorporates a GPRS Shield (General Packet Radio Services) that handles communication with a LAMP server (Linux, Apache, MySQL, PHP) connected to the Internet. It also incorporates a block of connection and disconnection of the electrical service that would leave the whole house without service. Two functionalities are used to present the data, one is local on the LCD display of the equipment installed in the home (user) and the second is remote access to a website (server). The results show that in comparison with a standard voltage device it presents an error of 0.28% and 4.12% in current. In this way, the use of this prototype for real-time monitoring of electricity consumption is validated, since it works similarly to a conventional one.

The increasingly anonymous methods people use to communicate in the modern world allow for more freedom of speech. The safety of anonymity, however, can enable criminals to cause harm to others through various means, such as blackmail, verbal abuse, threat letters and numerous other ways. These culprits, often hiding behind computer screens, can be extremely difficult to identify and especially difficult to find definitive proof of their wrongdoings. They are not completely untraceable, however, as they are bound to leave clues in the text, linking it to them. The way they phrase sentences, the words they use, how often they use them and other parts of their idiolect can be used to identify them and even connect them to other texts. Through analyzing the text, it becomes possible to catch these individuals. This analysis is neither simple nor cheap, the aid of linguistic experts is critical, and even they are likely to encounter difficulties. This article explores the way in which the work of such experts can be assisted through computer analysis based on machine learning techniques and the role Artificial Intelligence plays in bringing these criminals to justice. Our current paper investigates how linguistic features can be automatically extracted to be used in the field. Through a total of 61 real text artefacts written in the Hungarian language by four different individuals, we extract various syntactic and semantic linguistic features which reflect the author’s idiolect and aid the expert’s work. We demonstrate how the technique can aid author identification in criminology.

After evaluating the difficulty of CNNs in extracting convolution features, this paper suggested an improved convolutional neural network (CNN) method (ICNN-BNDOA), which is based on Batch Normalization (BN), Dropout (DO), and Adaptive Moment Estimation (Adam) optimizer. To circumvent the gradient challenge and quicken convergence, the ICNN-BNDOA uses a sequential CNN structure with the Leaky rectified linear unit (LeakyReLU) as the activation function (AF). The approach employs an Adam optimizer to handle the overfitting problem, which is done by introducing BN and DO layers to the entire connected CNN layers and the output layers, respectively, to decrease cross-entropy. Through a small regularization impact, BN was utilized to substantially speed up the training process of a neural network, as well as to increase the model's performance. The performance of the proposed system with conventional CNN (CCNN) was studied using the CIFAR-10 datasets as the benchmark data, and it was discovered that the suggested method demonstrated high recognition performance with the addition of BN and DO layers. CCNN and ICNN-BNDOA performance were compared. The statistical results showed that the proposed ICNN-BNDOA outperformed the CCNN with a training and testing accuracy of 0.6904 and 0.6861 respectively. It also outperformed with training and testing loss of 0.8910 and 0.9136 respectively.

Automatic analysis of financial data is the subject of many ongoing research. The aim of this study was to explore deep learning methods to predict and forecast the value of selected financial data included in financial quarterly reports based on historical data in a given future horizon, assuming incomplete data. The study considered the quarterly financial reports of selected companies listed on the Warsaw Stock Exchange (WSE), in the period from September 2010 to December 2021, where each report consists of about 250 indices. Basing on the principles of financial analysis and the interdependencies between financial indicators, a selection of interdependent indicators has been established to forecast future indices values based on historical data. The reinforced learning technique was used to see if it improves forecast results relative to the classic deep learning technique. Results indicate good prediction of the financial statement values up to one-year horizon, i.e., up to four future quarterly reports, considering both complete and incomplete financial data. Also, it is shown how observation update (reinforced learning) influences the forecast result. Forecast results are supporting tool for financial analysis for WSE.

An assertion is a Boolean expression embedded in a program that must hold during the execution. Executable assertions are a simple but practical way to check assumptions and code logic at runtime. Assertions are written by referring to concrete program states. In this paper, we recognize a variety of assertions that we call design assertions. These are assertions written to ensure design constraints and properties, not detailed implementation decisions, and thus can detect major problems in the implementation such as design drift or corrosion. However, they are written by referring to concrete program states, thus causing readability and maintenance problems. To address these problems, we propose to write design constraints at a higher abstraction level by referring to abstract program states. We explain our approach using the Dart/Flutter platform, but it should work in other languages and platforms with similar assertion facilities.

Data centers are critical environments for the availability of technology-based services. Aiming at the high availability of these services, performance metrics of nodes such as Virtual Machines (VM) or VMs clusters are widely monitored. These metrics, such as CPU and memory utilization, can show anomalous patterns associated with failures and performance degradation, culminating in resource exhaustion and total node failure. Thus, early detection of anomalies can enable remediation measures, such as VM migration and resource reallocation, before losses occur. However, traditional monitoring tools often use fixed thresholds for detecting problems on nodes and lack automatic ways to detect anomalies at runtime. In this sense, machine learning techniques have been reported to detect anomalies in computer systems with online and offline approaches. Thus, this work aims to propose and evaluate the DC Health application, pursues to anticipate the detection of anomalies in data center nodes. For this, this research was conducted from i) Systematic Literature Review previously performed, ii) problem modeling from real VM data and iii) DCH evaluation using the prequential method in 6 real-world datasets. Preliminary results showed that DCH excelled in constant memory usage and detection accuracy above 75% in the worst of the 5 cases and accuracy of 90% at best when applied in the real node datasets and an accuracy of 85% on the shuttle dataset. As a continuation of this research, it is expected to develop a case study with data center operators and the evaluation of the tool in a large volume of nodes.

The Internet has caused an evolution in how people socialize, work, and do business. The emergence and improvement in cloud computing and web technologies make interactions and remote processes possible. This advancement has presented an opportunity for the people and their representatives to meet during the voting process. Voting is making a choice or decision within a particular group. However, the conventional voting process that uses the paper-based approach faces the challenges of multiple voting, overvoting, cost, high voting fraud, and delay in declaring election results due to long counting times. Various methods have been proposed to overcome the multiple challenges prevalent in the traditional voting system. This paper proposes an antibot-based web voting platform that enables voters to vote within any location. It uses the hash technique and the antibot checking features to enforce security and voters’ confidentiality. PHP and HTML languages were used to implement the front-end of the system. SQL database and the Apache server were used for the back-end. On implementation and testing, our system shows good security enhancement and a reduction in the time consumed for counting and declaring election results.

The number of publicly accessible software repositories on online platforms is growing rapidly. With more than 128 million public repositories (as of March 2020), GitHub is the world’s largest platform for hosting and managing software projects. Where it used to be necessary to merge various data sources, it is now possible to access a wealth of data using the GitHub API alone. However, collecting and analyzing this data is not an easy endeavor. In this paper, we present Prometheus, a system for crawling and storing software repositories from GitHub. Compared to existing frameworks, Prometheus follows an event-driven microservice architecture. By separating functionality on the service level, there is no need to understand implementation details or use existing frameworks to extend or customize the system, only data. Prometheus consists of two components, one for fetching GitHub data and one for data storage which serves as a basis for future functionality. Unlike most existing crawling approaches, the Prometheus fetching service uses the GitHub GraphQL API. As a result, Prometheus can significantly outperform alternatives in terms of throughput in some scenarios.

Software requirement classification is crucial in segregating the user requirements into functional and quality requirements, based on their feedback or client demand. Doing so manually is time-consuming and not feasible. This can lead to delays in satisfying the requirements which in turn can lead to unhappier clients and users. Thus, machine learning techniques are used to optimize this task. In this work, five different word embedding techniques have been applied to the functional and non-functional (quality) software requirements. SMOTE is used to balance the numerical data obtained after word embedding. Dimensionality reduction and feature selection techniques are then employed to eliminate redundant and irrelevant features. Principal Component Analysis (PCA) is used for dimensionality reduction, and Rank-Sum test (RST) is used for feature selection. The resulting vectors are fed as inputs to eight different classifiers- Bagged k-Nearest Neighbors, Bagged Decision Tree, Bagged Naive-Bayes, Random Forest, Extra Tree, Adaptive Boost, Gradient Boosting, and a Majority Voting ensemble classifier, with Decision Tree, k-Nearest Neighbors, and Gaussian Naive Bayes. The experimental results suggest that the combination of word embedding and feature selection techniques with the various classifiers are successful in accurately classifying functional and quality software requirements.

Have you ever wondered why a song or a book or a movie becomes so popular that everyone everywhere starts talking about it? If we did not have the technology, we would say that people who love something would start recommending it to their friends and families. We live in the age of technology where there are so many algorithms that can discover the patterns of human interaction and make an excellent guess about someone’s opinion about something. These algorithms are building blocks of digital streaming services and E-Commerce websites. These services require as accurate as possible recommendation systems for them to function. While many businesses prefer one type or another of recommendation algorithms, in this study, we developed a hybrid recommendation system for a book E-Commerce website by integrating many popular classical and Deep Neural Network-based recommendation algorithms. Since explicit feedback is unavailable most of the time, all our implementations are on implicit binary feedback. The four algorithms that we were concerned about in this study were the well-known Collaborative filtering algorithms, item-based CF and user-based CF, ALS Matrix Factorization, and Deep Neural Network Based approaches. Consequently, comparing their performances and accuracy, it was not surprising that the Deep Neural Network approach was the most accurate recommender for our E-Commerce website.

Today, practically every bank’s computer system can automatically categorize transactions. If someone uses their debit/credit card to buy groceries or clothing, they can see the sort of expense on their user account in seconds. Even though banks provide this level of categorization for individual users, there is no categorization solution for accounting systems. In this article, the main objective is to design and develop a business workflow that can predict bank account transaction types. Various machine learning and deep learning algorithms are used to accomplish this purpose. In the prototype implementation, Support Vector Machines, Random Forest, Long Short-Term Memory Networks, and Frequent Pattern Growth algorithms are used, and the prediction successes of these techniques are analyzed.