Skip to main content

About this book

These transactions publish research in computer-based methods of computational collective intelligence (CCI) and their applications in a wide range of fields such as the semantic Web, social networks, and multi-agent systems. TCCI strives to cover new methodological, theoretical and practical aspects of CCI understood as the form of intelligence that emerges from the collaboration and competition of many individuals (artificial and/or natural). The application of multiple computational intelligence technologies, such as fuzzy systems, evolutionary computation, neural systems, consensus theory, etc., aims to support human and other collective intelligence and to create new forms of CCI in natural and/or artificial systems. This eighteenth issue contains 9 carefully selected and revised contributions.

Table of Contents


Using Semantic Web for Generating Questions: Do Different Populations Perceive Questions Differently?

In this paper, I propose an approach to using semantic web data for generating questions that are intended to help people develop arguments in a discussion session. Applying this approach, a question generation system that exploits WordNet for generating questions for argumentation has been developed. This paper describes a study that investigates a research question of whether different populations perceive questions (either generated by a system or by human experts) differently. To conduct this study, I asked eight human experts of the argumentation and the question generation communities to construct questions for three discussion topics and used a question generation system for generating questions for argumentation. Then, the author invited three groups of researchers to rate the mix of questions: (1) computer scientists, (2) researchers of the argumentation and question generation communities, and (3) student teachers for Computer Science. The evaluation study showed that human-generated questions were perceived differently by three different populations over three quality criteria (the understandability, the relevance, and the usefulness). For system-generated questions, the hypothesis could only be confirmed on the criteria of relevance and usefulness of questions. This contribution of the paper motivates researchers of question generation to deploy various techniques to generate questions adaptively for different target groups.
Nguyen-Thinh Le

Reflection of Intelligent E-Learning/Tutoring - The Flexible Learning Model in LMS Blackboard

The article encompasses the theoretical background and practical concept of teaching/learning through online courses as an example of smart solution of e-learning system adjusting to individual learning preferences, including students’ reflection on the individualized online instruction. First, the ‘Unlocking the Will to Learn’ concept by C.A. Johnston is introduced and implemented in the research design. Second, the pedagogical experiment reflecting individual learning style preferences is conducted. For this phase of research the e-application was designed which considers learners’ individual characteristics and consequently generates the online course content adjusted to them. Finally, learners’ feedback after studying in the course is presented.
Ivana Simonova, Petra Poulova, Pavel Kriz

GLIO: A New Method for Grouping Like-Minded Users

Grouping like-minded users is one of the emerging problems in Social Network Analysis. Indeed, it gives a good idea about group formation and social network evolution. Also, it explains various social phenomena and leads to many applications, such as friends suggestion and collaborative filtering. In this paper, we introduce a novel unsupervised method for grouping like-minded users within social networks. Such a method detects groups of users sharing the same interest centers and having similar opinions. In fact, the proposed method is based on extracting the interest centers and retrieving the polarities from the user’s textual posts. We validate our results by employing multiple clustering evaluation measures (recall, precision, F-score and Rand-Index). We compare our algorithm to a number of other clustering algorithms and opinion detection API. Results prove that the algorithm presented is efficient.
Soufiene Jaffali, Hanen Ameur, Salma Jamoussi, Abdelmajid Ben Hamadou

A Preferences Based Approach for Better Comprehension of User Information Needs

Within Mobile information retrieval research, context information provides an important basis for identifying and understanding user’s information needs. Therefore search process can take advantage of contextual information to enhance the query and adapt search results to user’s current context. However, the challenge is how to define the best contextual information to be integrated in search process. In this paper, our intention is to build a model that can identify which contextual dimensions strongly influence the outcome of the retrieval process and should therefore be in the user’s focus. In order to achieve these objectives, we create a new query language model based on user’s pereferences. We extend this model in order to define a relevance measure for each contextual dimension, which allow to automatically classify each dimension. This latter is used to compute the degree of change in result lists for the same query enhanced by different dimensions. Our experiments show that our measure can analyze the real user’s context of up to 12000 of dimensions (related to 4000 queries). We also show experimentally the quality of the set of contextual dimensions proposed, and the interest of the measure to understand mobile user’s needs and to enhance his query.
Sondess Missaoui, Rim Faiz

Performance Evaluation of the Customer Relationship Management Agent’s in a Cognitive Integrated Management Support System

The biggest problem currently, turns out to be the processing of unstructured knowledge in integrated management support systems. Note that knowledge contained in these systems is normally structuralized and the systems employ various methods for processing structuralized knowledge. However, in contemporary companies, unstructured knowledge is essential, mainly due to the possibility to obtain better flexibility and competitiveness of the organization. The users’ opinions about products can serve as example. Therefore, unstructured knowledge supports structuralized knowledge to a high degree. This paper presents the issues related to the sentiment analysis of customers’ opinions performed by Customer Relationship Management agent running in multi-agent Cognitive Integrated Management Information System. This system is an application of computational collective intelligence and allows for supporting the management processes related with all the domain of enterprise’s functioning. The agents are based on the Learning Intelligent Distribution Agent cognitive architecture, described shortly in the first part of the paper. Next, the logical architecture of Cognitive Integrated Management Information System are described. The main part of article presents issues related to functionality and implementation of Customer Relationship Management agent aims to sentiment analysis. The results of research experiment, aims to performance evaluation, are presented at the last part of article.
Marcin Hernes

Agreements Technologies - Towards Sophisticated Software Agents in Multi-agent Environments

Agreements are one of vital social concepts that help human agents to facilitate interactions in social settings. Nowadays, a challenging interdisciplinary scientific research includes all the processes and mechanisms concerned in reaching agreements between different kinds of agents. Enhancing agents with “social” abilities is newest trend in application of agent technology and the implementation of wide range of multi-agent systems.
Agreement Technologies bring a new flavor in implementation of more sophisticated autonomous software agents that mutually negotiate in order to come as much as close to win-win situation and to acceptable agreements.
The goal of the paper is to present key concepts in this area and highlight influence of Agreement Technologies on development of more sophisticated multi-agent systems. Several interesting systems and environments from different domains are briefly presented.
Mirjana Ivanović, Zoran Budimac

Identification of Underestimated and Overestimated Web Pages Using PageRank and Web Usage Mining Methods

The paper describes an alternative method of website analysis and optimization that combines methods of web usage and web structure mining - discovering of web users’ behaviour patterns as well as discovering knowledge from the website structure. Its primary objective is identifying of web pages, in which the value of their importance, estimated by the website developers, does not correspond to the real behaviour of the website visitors. It was proved before that the expected visit rate correlate with the observed visit rate of the web pages. Consequently, the expected probabilities of visiting of web pages by a visitor were calculated using the PageRank method and observed probabilities were obtained from the web server log files using the web usage mining method. The observed and expected probabilities were compared using the residual analysis. While the sequence rules analysis can only uncover the potential problem of web pages with higher visit rate, the proposed method of residual analysis can also consider other web pages with a smaller visit rate. The obtained results can be successfully used for a website optimization and restructuring, improving website navigation, and adaptive website realisation.
Jozef Kapusta, Michal Munk, Martin Drlík

Massive Classification with Support Vector Machines

The new boosting of Least-Squares SVM (LS-SVM), Proximal SVM (PSVM), Newton SVM (NSVM) algorithms aim at classifying very large datasets on standard personal computers (PCs). We extend the PSVM, LS-SVM and NSVM in several ways to efficiently classify large datasets. We developed a row incremental version for datasets with billions of data points. By adding a Tikhonov regularization term and using the Sherman-Morrison-Woodbury formula, we developed new algorihms to process datasets with a small number of data points but very high dimensionality. Finally, by applying boosting including AdaBoost and Arcx4 to these algorithms, we developed classification algorithms for massive, very-high-dimensional datasets. Numerical test results on large datasets from the UCI repository showed that our algorithms are often significantly faster and/or more accurate than state-of-the-art algorithms LibSVM, CB-SVM, SVM-perf and LIBLINEAR.
Thanh Nghi Do, Hoai An Le Thi

On a Multi-agent Distributed Asynchronous Intelligence-Sharing and Learning Framework

The current digital era is flooded with devices having high processing and networking capabilities. Sharing of information, learning and adaptation in such highly distributed systems can greatly enhance their performance and utility. However, achieving the same in the presence of asynchronous entities is a complex affair. Multi-agent system paradigms possess intrinsic similarities with these distributed systems and thus provide a fitting platform to solve the problems within. Traditional approaches to efficient information sharing and learning among autonomous agents in distributed environments incur high communication overheads. Non-conventional tactics based on social insect colonies provide natural solutions for transfer of social information in highly distributed and dense populations. This paper portrays a framework to achieve distributed and asynchronous sharing of intelligence and consequent learning among the entities of a networked distributed system. This framework couples localized communication with the available multi-agent technologies to realize asynchronous intelligence-sharing and learning. The framework takes in a user-defined objective together with a learning algorithm as inputs and facilitates cooperative learning among the agents using the mechanisms embedded within. The proposed framework has been implemented using Typhon agent framework over a LAN. The results obtained from the experiments performed using both static and dynamic LANs, substantiate the applicability of the proposed framework in real distributed mobile computing environments.
Shashi Shekhar Jha, Shivashankar B. Nair


Additional information

Premium Partner

    Image Credits