International Journal on Artificial Intelligence Tools

The present work presents a novel approach to song mood classification. Two language models, one absolute and one relative, are experimented with. Two distinct audio feature sets are compared against each other, and the significance of the inclusion of text stylistic features is established. Furthermore, Latent Semantic Analysis is innovatively combined with language modeling, and depicts the discriminative power of the latter. Finally, song “words” are defined in a broader sense that includes lyrics words as well as audio words, and LSA is applied to this augmented vocabulary with highly promising results. The methodology is applied to Greek songs, that are classified into one of four valence and into one of four arousal categories.

The advent and wide proliferation of Social Web in the recent years has promoted the concept of social interaction as an important influencing factor of the way enterprises and organizations conduct business. Among the fields influenced is that of Enterprise Knowledge Management, where the increase and maintenance of the employees' active participation in the organization's knowledge management activities is pursued through the adoption of social computing approaches. In this paper we consider a prominent and increasingly applied such approach, namely enterprise microblogging, and we propose a novel way to exploit its knowledge generation and sharing capabilities in order to effectively capture and formalize enterprise knowledge that is vague. Application and empirical evaluation of the framework indicates significant potential towards this goal.

A key challenge of adaptive interactive systems is to provide a positive user experience by extracting implicitly the users' unique characteristics through their interactions with the system, and dynamically adapting and personalizing the system's content presentation and functionality. Among the different dimensions of individual differences that could be considered, this work utilizes the cognitive styles of users as determinant factors for personalization. The overarching goal of this paper is to increase our understanding about the effect of cognitive styles of users on their navigation behavior and content representation preference. We propose a Web-based tool, utilizing Artificial Intelligence techniques, to implicitly capture and find any possible relations between the cognitive styles of users and their characteristics in navigation behavior and content representation preference by using their Web interaction data. The proposed tool has been evaluated with a user study revealing that cognitive styles of users have an effect on their navigation behavior and content representation preference. Research works like the reported one are useful for improving implicit and intelligent user modeling in engineering adaptive interactive systems.

This work studies collective intelligence behavior of Web users that share and watch video content. Accordingly, it is proposed that the aggregated users' video activity exhibits characteristic patterns. Such patterns may be used in order to infer important video scenes leading thus to collective intelligence concerning the video content. To this end, experimentation is based on users' interactions (e.g., pause, seek/scrub) that have been gathered in a controlled user experiment with information-rich videos. Collective information seeking behavior is then modeled by means of the corresponding probability distribution function. Thus, it is argued that the bell-shaped reference patterns are shown to significantly correlate with predefined scenes of interest for each video, as annotated by the users. In this way, the observed collective intelligence may be used to provide a video-segment detection tool that identifies the importance of video scenes. Accordingly, both a stochastic and a pattern matching approach are applied on the users' interactions information. The results received indicate increased accuracy in identifying the areas selected by users as having high importance information. In practice, the proposed techniques might improve both navigation within videos on the web as well as video search results with personalised video thumbnails.

Games have always been a popular test bed for artificial intelligence techniques. Game developers are always in constant search for techniques that can automatically create computer games minimizing the developer's task. In this work we present an evolutionary strategy based solution towards the automatic generation of two player board games. To guide the evolutionary process towards games, which are entertaining, we propose a set of metrics. These metrics are based upon different theories of entertainment in computer games. This work also compares the entertainment value of the evolved games with the existing popular board based games. Further to verify the entertainment value of the evolved games with the entertainment value of the human user a human user survey is conducted. In addition to the user survey we check the learnability of the evolved games using an artificial neural network based controller. The proposed metrics and the evolutionary process can be employed for generating new and entertaining board games, provided an initial search space is given to the evolutionary algorithm.

In many real-life datasets, the number of extracted frequent patterns was shown to be huge, hampering the effective exploitation of such amount of knowledge by human experts. To overcome this limitation, exact condensed representations were introduced in order to offer a small-sized set of elements from which the faithful retrieval of all frequent patterns is possible. In this paper, we introduce a new exact condensed representation only based on particular elements from the disjunctive search space. In this space, a pattern is characterized by its disjunctive support, i.e., the frequency of complementary occurrences – instead of the ubiquitous co-occurrence link – of its items. For several benchmark datasets, this representation has been shown interesting in compactness terms compared to the pioneering approaches of the literature. In this respect, we mainly focus here on proposing an efficient tool for mining this representation. For this purpose, we introduce an algorithm, called DSSRM, dedicated to this task. We also propose several techniques to optimize its mining time as well as its memory consumption. The carried out empirical study on benchmark datasets shows that DSSRM is faster by several orders of magnitude than the MEP algorithm.

Obtaining multi-objective optimization solutions with a small number of points smartly distributed along the Pareto front is a challenge. Optimization methods, such as the normalized normal constraint (NNC), propose the use of a filter to achieve a smart Pareto front distribution. The NCC optimization method presents several disadvantages related with the procedure itself, initial condition dependency, and computational burden. In this article, the epsilon-variable multi-objective genetic algorithm (ev-MOGA) is presented. This algorithm characterizes the Pareto front in a smart way and removes the disadvantages of the NNC method. Finally, examples of a three-bar truss design and controller tuning optimizations are presented for comparison purposes.