A neural network that matches with a complex data function is likely to boost the classification performance as it is able to learn the useful aspect of the highly varying data. In this work, the temporal context of the time series data is chosen as the useful aspect of the data that is passed through the network for learning. By exploiting the compositional locality of the time series data at each level of the network, shift-invariant features can be extracted layer by layer at different time scales. The temporal context is made available to the deeper layers of the network by a set of data processing operations based on the concatenation operation. A matching learning algorithm for the revised network is described in this paper. It uses gradient routing in the backpropagation path. The framework as proposed in this work attains better generalization without overfitting the network to the data, as the weights can be pretrained appropriately. It can be used end-to-end with multivariate time series data in their raw form, without the need for manual feature crafting or data transformation. Data experiments with electroencephalogram signals and human activity signals show that with the right amount of concatenation in the deeper layers of the proposed network, it can improve the performance in signal classification.
Due to the abundant amount of Customer's Reviews available in E-commerce platforms, Trust Reputation Systems remain reliable means to determine, circulate and restore the credibility and reputation of reviewers and their provided reviews. In fact before starting the process of Reputation score's calculation, we need to develop an accurate Sentiment orientation System able to extract opinion expressions, analyze them and determine the sentiment orientation of the Review and then classify it into positive, negative and objective. In this paper, we propose a novel semi-supervised approach which is a Combined Idiomatic-Ontology based Sentiment Orientation System (CIOSOS) that realizes a domain-dependent sentiment analysis of reviews. The main contribution of the system is to expand the general opinion lexicon SentiWordNet to a custom-made opinion lexicon (SentiWordNet++) with domain-dependent "opinion indicators" as well as "idiomatic expressions". The system relies also on a semi-supervised learning method that uses the general lexicon WordNet to identify synonyms or antonyms of the expanded terms and get their polarities from SentiWordNet and then store them in SentiWordNet++. The Sentiment polarity and the classification of the review provided by the CIOSOS is used as an input of our Reputation Algorithm proposed in previous papers in order to generate the Reputation score of the reviewer. We also provide an improvement in calculation method used to generate a "granular" reputation score of a feature or subfeature of the product.
<p class="0papertitle">In both the private and public sectors, human resource management processes face considerable challenges in terms of skills management within organizations. In fact, during the recruitment process, it is difficult to find the right profile for certain functions. To cope with this constraint and thus streamline this process, organizations tend to implement intelligent management of jobs and skills. Most systems of matching a job with a profile face the difficulty of developing and maintaining resources specific to each field. In view of this, ontologies are not only a tool for professional management and strategic management of human resources, but they also make it possible to base the relationship between the couple job / profile. Thus, we propose a construction approach of three ontologies that will play a key role in knowledge management in the context of the Secrétariat Chargé De L’eau but which remains valid for later use in broader contexts.</p>
This article proposes in depth comparative study of the most popular, used and analyzed Trust and Reputation System (TRS) according to the trust and reputation literature and in terms of specific trustworthiness criteria. This survey is realized relying on a selection of trustworthiness criteria that analyze and evaluate the maturity and effectiveness of TRS. These criteria describe the utility, the usability, the performance and the effectiveness of the TRS. We also provide a summary table of the compared TRS within a detailed and granular selection of trust and reputation aspects.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.