Artificial Intelligence Applications and Innovations

The categorical ethos of artificial intelligence is influenced by its basic structure, which defines its due purpose as a legal personality, challenging the conventional standards of law and justice in a globalized world. Recent developments show a precedential growth in the need-perspective of the AI industry, thereby influencing governance and corporate operations and their legal side in cross-cultural avenues. The determinant outlining of artificial intelligence as a legal personality rests on its probabilistic nature, which yet can be limited to the jurisprudential scope of AI-based on the ethos of the utilitarian approach involving the anthropocentric innovations for artificial intelligence. The dynamic nature of AI, however, in the proposition, is capable of a full-fledged and anthropomorphic legal representation and interpretation, which is hard to find in D9 and certain developing countries, which poses special risks to the generic legal infrastructure of a democratic polity to understand the dynamic and self-transformative nature of artificial intelligence in the age of globalization.The paper is thus based on the proposition that the ethos involving the legal infrastructure and persona of artificial intelligence is traceable and easier in deterministic mechanisms by regarding and extending stable & constitutive approaches to dissect the legal challenges connected with the redemptions implicated with the lack of a full-fledged regard and scope of the legal personality of AI. The approaches in due proposition are (a) anthropomorphisation; (b) naturalization; (c) techno-socialization; and (d) enculturation. Further, the paper analyses on the challenges to determine the problematic implications awaited by the influence of populism, protectionism, data-centred digital colonialism and technology distancing and proposes suggestions based on the four approaches to counter the minimal effects of the implications. The conclusions of the paper rest on the argument that in the case of a post-liberal order, the ethos of AI can be protected and diversified by adapting with the appreciation of the ethos of globalization, giving adequate, constitutive and reasonable space to the identity-led implications of national identity & diluting the monopolistic influence of the utilitarian approach to artificial intelligence.

EU Draft Ethics guidelines for Trustworthy AI [8] has been proposed to promote ethical, lawful and robust AI solutions. In this article, we entertain the systemic issues and challenges of any development of the proposed guidelines.

The performance of solving a constraint problem can often be improved by converting a subproblem into a single constraint (for example into a regular membership constraint or a table constraint). In the past, it stood out, that specialist constraint solvers (like simplex solver or SAT solver) outperform general constraint solvers, for the problems they can handle. The disadvantage of such specialist constraint solvers is that they can handle only a small subset of problems with special limitations to the domains of the variables and/or to the allowed constraints. In this paper we introduce the concept of fd-complete constraints and fd-complete constraint satisfaction problems, which allow combining both previous approaches. More accurately, we convert general constraint problems into problems which use only one, respectively one kind of constraint. The goal is it to interpret and solve the converted constraint problems with specialist solvers, which can solve the transformed constraint problems faster than the original solver the original constraint problems.

Advancements in Long Short Term Memory (LSTM) Networks have shown remarkable success in various Natural Language Generation (NLG) tasks. However, generating sequence from pre-specified lexical constraints is a new, challenging and less researched area in NLG. Lexical constraints take the form of words in the language model’s output to create fluent and meaningful sequences. Furthermore, most of the previous approaches cater this problem by allowing the inclusion of pre-specified lexical constraints during the decoding process, which increases the decoding complexity exponentially or linearly with the number of constraints. Moreover, some of the previous approaches can only deal with single constraint. Additionally, most of the previous approaches only deal with single constraints. In this paper, we propose a novel neural probabilistic architecture based on backward-forward language model and word embedding substitution method that can cater multiple lexical constraints for generating quality sequences. Experiments shows that our proposed architecture outperforms previous methods in terms of intrinsic evaluation.

This paper analyzes the impact of reservoir computing, and, in particular, of Deep Echo State Networks, to the modeling of highly non-linear dynamical systems that can be commonly found in the industry. Several applications are presented focusing on forecasting models related to energy content of steelwork byproduct gasses. Deep Echo State Network models are trained, validated and tested by exploiting datasets coming from a real industrial context, with good results in terms of accuracy of the predictions.

Deep Reinforcement Learning (DRL) is increasingly used to train robots to perform complex and delicate tasks, while the development of realistic simulators contributes to the acceleration of research on DRL for robotics. However, it is still not straightforward to employ such simulators in the typical DRL pipeline, since their steep learning curve and the enormous amount of development required to interface with DRL methods significantly restrict their use by researchers. To overcome these limitations, in this work we present an open-source framework that combines an established interface used by DRL researchers, the OpenAI Gym interface, with the state-of-the-art Webots robot simulator in order to provide a standardized way to employ DRL in various robotics scenarios. Deepbots aims to enable researchers to easily develop DRL methods in Webots by handling all the low-level details and reducing the required development effort. The effectiveness of the proposed framework is demonstrated through code examples, as well as using three use cases of varying difficulty.

A language independent deep learning (DL) architecture for machine translation (MT) evaluation is presented. This DL architecture aims at the best choice between two MT (S1, S2) outputs, based on the reference translation (Sr) and the annotation score. The outputs were generated from a statistical machine translation (SMT) system and a neural machine translation (NMT) system. The model applied in two language pairs: English - Greek (EN-EL) and English - Italian (EN-IT). In this paper, a variety of experiments with different parameter configurations is presented. Moreover, linguistic features, embeddings representation and natural language processing (NLP) metrics (BLEU, METEOR, TER, WER) were tested. The best score was achieved when the proposed model used source segments (SSE) information and the NLP metrics set. Classification accuracy has increased up to 5% (compared to previous related work) and reached quite satisfactory results for the Kendall τ score.

The recent advancements in the areas of computer vision and deep learning with the development of convolutional neural networks and the profusion of highly accurate general purpose pre-trained models, create new opportunities for the interaction of humans with systems and facilitate the development of advanced features for all types of platforms and applications. Research, consumer and industrial applications increasingly integrate deep learning frameworks into their operational flow, and as a result of the availability of high performance hardware (Computer Boards, GPUs, TPUs) also for individual consumers and home use, this functionality has been moved closer to the end-users, at the edge of the network. In this work, we exploit the aforementioned approaches and tools for the development of an edge-native platform for exergames, which includes innovative gameplay and features for the users. A prototype game was created using the platform that was deployed in the real-world scenario of a rehabilitation center. The proposed approach provides advanced user experience based on the automated, real-time pose and gesture detection, and in parallel maintains low-cost to enable wide adoption in multiple applications across domains and usage scenarios.

In last decade, cryptocurrency has emerged in financial area as a key factor in businesses and financial market opportunities. Accurate predictions can assist cryptocurrency investors towards right investing decisions and lead to potential increased profits. Additionally, they can also support policy makers and financial researchers in studying cryptocurrency markets behavior. Nevertheless, cryptocurrency price prediction is considered a very challenging task, due to its chaotic and very complex nature. In this study we evaluate some of the most successful and widely used deep learning algorithms forecasting cryptocurrency prices. The results obtained, provide significant evidence that deep learning models are not able to solve this problem efficiently and effectively. Conducting detailed experimentation and results analysis, we conclude that it is essential to invent and incorporate new techniques, strategies and alternative approaches such as: more sophisticated prediction algorithms, advanced ensemble methods, feature engineering techniques and other validation metrics.

Neural Architecture Search is becoming an increasingly popular research field and method to design deep learning architectures. Most research focuses on searching for small blocks of deep learning operations, or micro-search. This method yields satisfactory results but demands prior knowledge of the macro architecture’s structure. Generally, methods that do not utilize macro structure knowledge perform worse but are able to be applied to datasets of completely new domains. In this paper, we propose a macro NAS methodology which utilizes concepts of Regularized Evolution and Macro Neural Architecture Search (DeepNEAT), and apply it to the Fashion-MNIST dataset. By utilizing our method, we are able to produce networks that outperform other macro NAS methods on the dataset, when the same post-search inference methods are used. Furthermore, we are able to achieve 94.46% test accuracy, while requiring considerably less epochs to fully train our network.

The field of sound event detection is a growing sector which has mainly focused on the identification of sound classes from daily life situations. In most cases these sound detection models are trained on publicly available sound databases, up to now, however, they do not include acoustic data from manufacturing environments. Within manufacturing industries, acoustic data can be exploited in order to evaluate the correct execution of assembling processes. As an example, in this paper the correct plugging of connectors is analyzed on the basis of multimodal contextual process information. The latter are the connector’s acoustic properties and visual information recorded in form of video files while executing connector locking processes.For the first time optical microphones are used for the acquisition and analysis of connector sound data in order to differentiate connector locking sounds from each other respectively from background noise and sound events with similar acoustic properties. Therefore, different types of feature representations as well as neural network architectures are investigated for this specific task.The results from the proposed analysis show, that multimodal approaches clearly outperform unimodal neural network architectures for the task of connector locking validation by reaching maximal accuracy levels close to 85$$\%$$. Since in many cases there are no additional validation methods applied for the detection of correctly locked connectors in manufacturing industries, it is concluded that the proposed connector lock event detection framework is a significant improvement for the qualitative validation of plugging operations.

This work addresses the problem of helping speech therapists in interpreting results of tachistoscopes. These are instruments widely employed to diagnose speech and reading disorders. Roughly speaking, they work as follows. During a session, some strings of letters, which may or not correspond to existing words, are displayed to the patient for an amount of time set by the therapist. Next, the patient is asked for typing the read string. From the machine learning point of view, this raise an interesting problem of analyzing the sets of input and output words to evaluate the presence of a pathology.

The South Asian countries have the most polluted cities in the world which has caused quite a concern in the recent years due to the detrimental effect it had on economy and on health of humans and crops. PM 2.5 in particular has been linked to cardiovascular diseases, pulmonary diseases, increased risk of lung cancer and acute respiratory infections. Higher concentration of surface ozone has been observed to have negatively impacted agricultural yield of crops. Due to its deleterious impact on human health and agriculture, air pollution cannot be brushed off as a trivial matter and measures must be taken to address the problem. Deterministic models have been actively used; but they fall short due to their complexity and inability to accurately model the problem. Deep learning models have however shown potential when it comes to modeling time series data. This article explores the use of recurrent neural networks as a framework for predicting the hazard levels in Lahore, Pakistan with 95.0% accuracy and Beijing, China with 98.95% using the time series data of air pollutants and meteorological parameters. Forecasting air quality index (AQI) and Hazard levels would help the government take appropriate steps to enact policies to reduce the pollutants and keep the citizens informed about the statistics.

In fuzzy hypothesis testing we use fuzzy test statistics produced by fuzzy estimators and fuzzy critical values. In this paper we use the non-asymptotic fuzzy estimators in fuzzy hypothesis testing. These are triangular shaped fuzzy numbers that generalize the fuzzy estimators based on confidence intervals in such a way that eliminates discontinuities and ensures compact support. Our approach is particularly useful in critical situations, where subtle fuzzy comparisons between almost equal statistical quantities have to be made. In such cases the hypotheses tests that use non-asymptotic fuzzy estimators give better results than the previous approaches, since they give us the possibility of partial rejection or not of $$H_0$$.

In this paper, we investigate a special case of an open problem that is related to the exchange principle, a property of fuzzy implications. We focus on the cases of (S,N)– implications and the preservation of the exchange principle via lattice operations. We present and prove some sufficient conditions such that the exchange principle is preserved under the join and meet operations if we use (S,N)– implications.

The increase of the adoption of IoT devices and the contemporary problem of food production have given rise to numerous applications of IoT in agriculture. These applications typically comprise a set of sensors that are installed in open fields and measure metrics, such as temperature or humidity, which are used for irrigation control systems. Though useful, most contemporary systems have high installation and maintenance costs, and they do not offer automated control or, if they do, they are usually not interpretable, and thus cannot be trusted for such critical applications. In this work, we design Vital, a system that incorporates a set of low-cost sensors, a robust data store, and most importantly an explainable AI decision support system. Our system outputs a fuzzy rule-base, which is interpretable and allows fully automating the irrigation of the fields. Upon evaluating Vital in two pilot cases, we conclude that it can be effective for monitoring open-field installations.

In this paper, acoustic resonance testing on glass intravenous (IV) bottles is presented. Different machine learning methods were applied to distinguish acoustic observations of bottles with defects from the intact ones. Due to the very limited amount of available specimens, the question arises whether the deep learning methods can achieve similar or even better detection performance compared with traditional methods.The results from the binary classification experiments are presented and compared in terms of Balanced Accuracy Rate, F1-score, Area Under the Receiver Operating Characteristic Curve and Matthews Correlation Coefficient metrics.The presented feature analysis and the employed classifiers achieved solid results, despite the rather small and imbalanced dataset with a highly inconsistent class population.

Supervisory Control and Data Acquisition (SCADA) systems used in wind turbines for monitoring the health and performance of a wind farm can suffer from data loss due to sensor failure, transmission link breakdown or network congestion. Sensory data is used for important control decisions and such data loss can make the failures harder to detect. This work proposes various solutions to reconstruct the lost information of important SCADA parameters using Linear and non-linear Artificial Intelligence (AI) algorithms. It comprises of three major contributions; (1) signal reconstruction from other available SCADA parameters, (2) comparison of linear and non-linear AI models, and (3) generalization of the AI algorithms between turbines. Experimental results demonstrate the effectiveness of the developed methodologies for reconstruction of the lost information for valuable planning decisions.

Drone swarms are becoming a new tool for many tasks including surveillance, search, rescue, construction, and defense related activities. As their usage increases, so does the possibility of adversarial attacks on their contribution to these use cases. One possible avenue, whether deliberate or not, is to deny access to the position feedback offered by the Global Positioning System (GPS). Operating in these ‘GPS denied’ environments poses a new challenge; both in navigation, and in collision avoidance. This study proposes two novel concepts; a structural model of environmental deviance to aid in autonomous navigation, and a method to use the output of said model to implement a collision avoidance system. Both of these concepts are developed and tested in the framework of a simulated environment that mimics a GPS-denied scenario. Using data from hundreds of simulated swarm flights, this work shows structured learning can improve navigational accuracy without the need for externally provided position feedback.

This paper presents a two-stage Machine Learning (ML) model to predict the arrival time of In-Process Control (IPC) samples at the quality testing laboratories of a chemical company. The model was developed using three iterations of the CRoss-Industry Standard Process for Data Mining (CRISP-DM) methodology, each focusing on a different regression approach. To reduce the ML analyst effort, an Automated Machine Learning (AutoML) was adopted during the modeling stage of CRISP-DM. The AutoML was set to select the best among six distinct state-of-the-art regression algorithms. Using recent real-world data, the three main regression approaches were compared, showing that the proposed two-stage ML model is competitive and provides interesting predictions to support the laboratory management decisions (e.g., preparation of testing instruments). In particular, the proposed method can accurately predict 70% of the examples under a tolerance of 4 time units.

This paper presents a novel Machine Learning (ML) approach to support the creation of woven fabrics. Using data from a textile company, two CRoss-Industry Standard Process for Data Mining (CRISP-DM) iterations were executed, aiming to compare three input feature representation strategies related with fabric design and finishing processes. During the modeling stage of CRISP-DM, an Automated ML (AutoML) procedure was used to select the best regression model among six distinct state-of-the-art ML algorithms. A total of nine textile physical properties were modeled (e.g., abrasion, elasticity, pilling). Overall, the simpler yarn representation strategy obtained better predictive results. Moreover, for eight fabric properties (e.g., elasticity, pilling) the addition of finishing features improved the quality of the predictions. The best ML models obtained low predictive errors (from 2% to 7%) and are potentially valuable for the textile company, since they can be used to reduce the number of production attempts (saving time and costs).

Surfing is currently one of the most popular water sports in the world, both for recreational and competitive level surfers. Surf session analysis is often performed with commercially available devices. However, most of them seem insufficient considering the surfers’ needs, by displaying a low number of features, being inaccurate, invasive or not adequate for all surfer levels. Despite the fact that performing manoeuvres is the ultimate goal of surfing, there are no available solutions that enable the identification and characterization of such events. In this work, we propose a novel method to detect manoeuvre events during wave riding periods resorting solely to the inertial sensors embedded in smartphones. The proposed method was able to correctly identify over 95% of all the manoeuvres in the dataset (172 annotated manoeuvres), while achieving a precision of up to 80%, using a session-independent validation approach. These findings demonstrate the suitability and validity of the proposed solution for identification of manoeuvre events in real-world conditions, evidencing a high market potential.

Internet of Things (IoT) devices are inherently vulnerable due to insecure design, implementation, and configuration. Aggressive behavior change, due to increased attacker’s sophistication, and the heterogeneity of the data in IoT have proven that securing IoT devices is a making challenge. To detect intensive attacks and increase device uptime, we propose a novel ensemble learning model for IoT anomaly detection using software-defined networks (SDN). We use a deep auto-encoder to extract handy features for stacking into an ensemble learning model. The learned model is deployed in the SDN controller to detect anomalies or dynamic attacks in IoT by addressing the class imbalance problem. We validate the model with real-time testbed and benchmark datasets. The initial results show that our model has a better and more reliable performance than the competing models showcased in the relevant related work.

Twitter is generating a large amount of real-time data in the form of microblogs that has potential knowledge for various applications like traffic incident analysis and urban planning. Social media data represents the unbiased actual insights of citizens’ concerns that may be mined in making cities smarter. In this study, a computational framework has been proposed using word embedding and machine learning model to detect traffic incidents using social media data. The study includes the feasibility of using machine learning algorithms with different feature extraction and representation models for the identification of traffic incidents from the Twitter interactions. The comprehensive proposed approach is the combination of following four steps. In the first phase, a dictionary of traffic-related keywords is formed. Secondly, real-time Twitter data has been collected using the dictionary of identified traffic related keywords. In the third step, collected tweets have been pre-processed, and the feature generation model is applied to convert the dataset eligible for a machine learning classifier. Further, a machine learning model is trained and tested to identify the tweets containing traffic incidents. The results of the study show that machine learning models built on top of right feature extraction strategy is very promising to identify the tweets containing traffic incidents from micro-blogs.

The therapy of patients with psychotic disorders (i.e., bipolar disorder and schizophrenia) could benefit from the constant monitoring of their physiological and motor parameters. In this paper, we present an innovative and advanced cloud based platform that facilitates the effective monitoring of such patients. A commodity smartwatch is used for biosignal and motion data collection at a 24/7 basis. The paper describes the technical details of the implemented application both on the smartwatch and the cloud server side. Technical challenges regarding the upload, the storage and the battery constraints of the smartwatch are also discussed, along with the initial results regarding data visualization and processing.

Dementia has a large negative impact on the global healthcare and society. Diagnosis is rather challenging as there is no standardised test. The purpose of this paper is to conduct an analysis on ADNI data and determine its effectiveness for building classification models to differentiate the categories Cognitively Normal (CN), Mild Cognitive Impairment (MCI), and Dementia (DEM), based on tuning three Deep Learning models: two Multi-Layer Perceptron (MLP1 and MLP2) models and a Convolutional Bidirectional Long Short-Term Memory (ConvBLSTM) model. The results show that the MLP1 and MLP2 models accurately distinguish the DEM, MCI and CN classes, with accuracies as high as 0.86 (SD 0.01). The ConvBLSTM model was slightly less accurate but was explored in view of comparisons with the MLP models, and for future extensions of this work that will take advantage of time-related information. Although the performance of ConvBLSTM model was negatively impacted by a lack of visit code data, opportunities were identified for improvement, particularly in terms of pre-processing.

Despite the success of Convolutional Neural Network-based Computer-Aided Diagnosis research, its clinical applications remain challenging. Accordingly, developing medical Artificial Intelligence (AI) fitting into a clinical environment requires identifying/bridging the gap between AI and Healthcare sides. Since the biggest problem in Medical Imaging lies in data paucity, confirming the clinical relevance for diagnosis of research-proven image augmentation techniques is essential. Therefore, we hold a clinically valuable AI-envisioning workshop among Japanese Medical Imaging experts, physicians, and generalists in Healthcare/Informatics. Then, a questionnaire survey for physicians evaluates our pathology-aware Generative Adversarial Network (GAN)-based image augmentation projects in terms of Data Augmentation and physician training. The workshop reveals the intrinsic gap between AI/Healthcare sides and solutions on Why (i.e., clinical significance/interpretation) and How (i.e., data acquisition, commercial deployment, and safety/feeling safe). This analysis confirms our pathology-aware GANs’ clinical relevance as a clinical decision support system and non-expert physician training tool. Our findings would play a key role in connecting inter-disciplinary research and clinical applications, not limited to the Japanese medical context and pathology-aware GANs.

Long term stress often causes depression and neuronal atrophies that in turn can lead to a variety of health problems. As a result of these cellular changes, also molecular changes occur. These changes, that include increase of glucocorticoids and decrease of the brain-derived neurotrophic factor, have the unfortunate effect that they decrease the cognitive abilities needed for the individual to solve the stressful situation. Such cognitive abilities like reappraisal and their adaptation mechanisms turn out to be substantially impaired while they are needed for regulation of the negative emotions. However, antidepressant treatments and some other therapies have proved to be quite effective for the strengthening of such cognitive abilities. This study introduces an adaptive causal network model for this phenomenon where a subject loses his or her cognitive abilities (negative metaplasticity) due to long-term stress and re-improve these cognitive abilities (positive metaplasticity) through mindfulness-based cognitive therapy (MBCT). Simulation results have been reported for demonstration of the phenomenon.

Cancer, which involves the dysregulation of genes via multiple mechanisms, is unlikely to be fully explained by a single data type. By combining different “omes”, researchers can increase the discovery of novel bio-molecular associations with disease-related phenotypes. Investigation of functional relations among genes associated with the same disease condition may further help to develop more accurate disease-relevant prediction models. In this work, we present an integrative framework called Data & Analytic Integrator (DAI), to explore the relationship between different omics via different mathematical formulations and algorithms. In particular, we investigate the combinatorial use of molecular knowledge identified from omics integration methods netDx, iDRW and SSL, by fusing the derived aggregated similarity matrices and by exploiting these in a semi-supervised learner. The analysis workflows were applied to real-life data for ovarian cancer and underlined the benefits of joint data and analytic integration.

Early diagnosis of some life-threatening diseases such as cancers and heart is crucial for effective treatments. Supervised machine learning has proved to be a very useful tool to serve this purpose. Historical data of patients including clinical and demographic information is used for training learning algorithms. This builds predictive models that provide initial diagnoses. However, in the medical domain, it is common to have the positive class under-represented in a dataset. In such a scenario, a typical learning algorithm tends to be biased towards the negative class, which is the majority class, and misclassify positive cases. This is known as the class imbalance problem. In this paper, a framework for predictive diagnostics of diseases with imbalanced records is presented. To reduce the classification bias, we propose the usage of an overlap-based undersampling method to improve the visibility of minority class samples in the region where the two classes overlap. This is achieved by detecting and removing negative class instances from the overlapping region. This will improve class separability in the data space. Experimental results show achievement of high accuracy in the positive class, which is highly preferable in the medical domain, while good trade-offs between sensitivity and specificity were obtained. Results also show that the method often outperformed other state-of-the-art and well-established techniques.

The use of chatbots evolved rapidly in numerous fields in recent years, including Marketing, Supporting Systems, Education, Health Care, Cultural Heritage, and Entertainment. In this paper, we first present a historical overview of the evolution of the international community’s interest in chatbots. Next, we discuss the motivations that drive the use of chatbots, and we clarify chatbots’ usefulness in a variety of areas. Moreover, we highlight the impact of social stereotypes on chatbots design. After clarifying necessary technological concepts, we move on to a chatbot classification based on various criteria, such as the area of knowledge they refer to, the need they serve and others. Furthermore, we present the general architecture of modern chatbots while also mentioning the main platforms for their creation. Our engagement with the subject so far, reassures us of the prospects of chatbots and encourages us to study them in greater extent and depth.

Nowadays, Intelligent Personal Agents include Natural Language Understanding (NLU) modules, that utilize Machine Learning (ML), which can be included in different kind of applications in order to enable the translation of users’ input into different kinds of actions, as well as ML modules that handle dialogue. This translation is attained by the matching of a user’s sentence with an intent contained in an Agent. This paper introduces the first generation of the CERTH Intelligent Personal Agent (CIPA) which is based on the RASA ( https://rasa.com/ ) framework and utilizes two machine learning models for NLU and dialogue flow classification. Besides the architecture of CIPA—Generation A, a novel dialogue-story generator that is based on the idea of adjacency pairs is introduced. By utilizing on this novel-generator, the agent is able to create all the possible dialog trees in order to handle conversations without training on existing data in contrast with the majority of the current alternative solutions. CIPA supports multiple intents and it is capable of classifying complex sentences consisting of two user’s intents into two automatic operations from the part of the agent. The introduced CIPA—Generation A has been deployed and tested in a real-world scenario at Centre’s of Research & Technology Hellas (CERTH) nZEB Smart Home ( https://smarthome.iti.gr/ ) in two different domains, energy and health domain.

Question Answering (QA) and Question Generation (QG) have been subjects of an intensive study in recent years and much progress has been made in both areas. However, works on combining these two topics mainly focus on how QG can be used to improve QA results. Through existing Natural Language Processing (NLP) techniques, we have implemented a tool that addresses these two topics separately. We further use them jointly in a pipeline. Thus, our goal is to understand how these modules can help each other. For QG, our methodology employs a detailed analysis of the relevant content of a sentence through Part-of-speech (POS) tagging and Named Entity Recognition (NER). Ensuring loose coupling with the QA task, in the latter we use Information Retrieval to rank sentences that might contain relevant information regarding a certain question, together with Open Information Retrieval to analyse the sentences. In its current version, the QG tool takes a sentence to formulate a simple question. By connecting QG with the QA component, we provide a means to effortlessly generate a test set for QA. While our current QA approach shows promising results, when enhancing the QG component we will, in the future, provide questions for which a more elaborated QA will be needed. The generated QA datasets contribute to QA evaluation, while QA proves to be an important technique for assessing the ambiguity of the questions.

The automation of fake news detection is the focus of a great deal of scientific research. With the rise of social media over the years, there has been a strong preference for users to be informed using their social media account, leading to a proliferation of fake news through them. This paper evaluates the veracity of politically-oriented news and in particular the tweets about the recent event of Hong Kong protests, with the aid of a dataset recently published by Twitter. From this dataset, Chinese tweets are translated into English, which are kept along with originally English tweets. By utilizing a language-independent filtering process, relevant tweets are identified. To complete the dataset, tweets originating from valid sources are used as the real portion, with journalists rather than news agencies being considered, which constitutes a novel aspect of the methodology. Well-known Machine Learning algorithms are used to classify tweets, which are represented by a feature value vector that is extracted, selected and preprocessed from the datasets and mainly revolves around language use, with word entropy being a novel feature. The results derived from these algorithms highlight morphological, lexical and vocabulary differences between tweets spreading fake and real news, which are for the most part in accordance with past related work.

Word Embeddings are used widely in multiple Natural Language Processing (NLP) applications. They are coordinates associated with each word in a dictionary, inferred from statistical properties of these words in a large corpus. In this paper we introduce the notion of “concept” as a list of words that have shared semantic content. We use this notion to analyse the learnability of certain concepts, defined as the capability of a classifier to recognise unseen members of a concept after training on a random subset of it. We first use this method to measure the learnability of concepts on pretrained word embeddings. We then develop a statistical analysis of concept learnability, based on hypothesis testing and ROC curves, in order to compare the relative merits of various embedding algorithms using a fixed corpora and hyper parameters. We find that all embedding methods capture the semantic content of those word lists, but fastText performs better than the others.

Nowadays, the fashion industry is moving towards fast fashion, offering a large selection of garment products in a quicker and cheaper manner. To this end, the fashion designers are required to come up with a wide and diverse amount of fashion products in a short time frame. At the same time, the fashion retailers are oriented towards using technology, in order to design and provide products tailored to their consumers’ needs, in sync with the newest fashion trends. In this paper, we propose an artificial intelligence system which operates as a personal assistant to a fashion product designer. The system’s architecture and all its components are presented, with emphasis on the data collection and data clustering subsystems. In our use case scenario, datasets of garment products are retrieved from two different sources and are transformed into a specific format by making use of Natural Language Processes. The two datasets are clustered separately using different mixed-type clustering algorithms and comparative results are provided, highlighting the usefulness of the clustering procedure in the clothing product recommendation problem.

This paper documents the efforts in implementing lyric generation machine learning models in the Greek language for the genre of Éntekhno music. To accomplish this, we used three different Long Short-Term Memory Recurrent Neural Network approaches. The first method utilizes word-level bi-directional network models, the second method expands on the first by learning the word embeddings on the initial layer of the network, while the last method is based on a char-level network model. Our experimental procedure, which utilized a high sample of human judges, shows that texts of lyrics generated by our models are of high quality and are not that easily distinguishable from actual lyrics.