Search results

The purpose of this paper is to propose an automatic method to generate semantic annotations of football transfer in the news. The current automatic news integration systems on the Web are constantly faced with the challenge of diversity, heterogeneity of sources. The approaches for information representation and storage based on syntax have some certain limitations in news searching, sorting, organizing and linking it appropriately. The models of semantic representation are promising to be the key to solving these problems.

The approach of the author leverages Semantic Web technologies to improve the performance of detection of hidden annotations in the news. The paper proposes an automatic method to generate semantic annotations based on named entity recognition and rule-based information extraction. The authors have built a domain ontology and knowledge base integrated with the knowledge and information management (KIM) platform to implement the former task (named entity recognition). The semantic extraction rules are constructed based on defined language models and the developed ontology.

The proposed method is implemented as a part of the sport news semantic annotations-generating prototype BKAnnotation. This component is a part of the sport integration system based on Web Semantics BKSport. The semantic annotations generated are used for improving features of news searching – sorting – association. The experiments on the news data from SkySport (2014) channel showed positive results. The precisions achieved in both cases, with and without integration of the pronoun recognition method, are both over 80 per cent. In particular, the latter helps increase the recall value to around 10 per cent.

This is one of the initial proposals in automatic creation of semantic data about news, football news in particular and sport news in general. The combination of ontology, knowledge base and patterns of language model allows detection of not only entities with corresponding types but also semantic triples. At the same time, the authors propose a pronoun recognition method using extraction rules to improve the relation recognition process.

The purpose of this paper is to enhance the quality of new reducing sentence in sentence-generation-based summarizing method by establishing consequence relationship between two action, state or process Vietnamese sentences.

First, types of pairs of Vietnamese sentences based on presupposition about the consequence relationship is classified: the verb indicating action or state at the first sentence is considered as the consequence of the verb indicating action, state or process at the second sentence. Then main predicates in Discourse Representation Structure – a logical form which represents the semantic of a given pair of sentences – is analyzed and inner- and inter-sentential relationships are determined. The next step is to generate the syntactic structure of the new reducing sentence. Finally, a combination with the built set of lexicons is done to complete the new meaning-summarizing Vietnamese sentence.

This method makes the new meaning-summarizing Vietnamese sentence satisfy two requirements: summarize the semantic of the given pair of Vietnamese sentences and have naturalism in common Vietnamese communication. In addition, it is possible to extend the method and apply for the purpose of summarizing the more complex Vietnamese paragraphs as well as paragraphs in other languages.

At the first step, only inter-sentential consequence relationship is considered and this is applied to the limit types of pairs of Vietnamese sentences which have a simple structure.

This study presents improvements in sentence-generation-based summarization method to enhance the quality of new meaning-summarizing Vietnamese sentences. This method proves effective in summarizing the considered pairs of sentences.

The purpose of this paper is to activate latent users posts by modeling user behaviors by a transition of clusters that represent particular posting activities. Twitter has rapidly spread and become an easy and convenient microblog that enables users to exchange instant text messages called tweets. There are so many latent users whose posting activities have decreased.

Under this model, two kinds of time-series analysis methods are proposed to clarify the lifecycles of Twitter users. In the first one, all users belong to a cluster consisting of several features at individual time slots and move among the clusters in a time series. In the second one, the posting activities of Twitter users are analyzed by the amount of tweets that vary with time.

This sophisticated evaluation using a large actual tweet-set demonstrated the proposed methods effectiveness. The authors found a big difference in the state transition diagrams between long- and short-term users. Analysis of short-term users introduces effective knowledge for encouraging continued Twitter use.

An the efficient user behavior model, which describes transitions of posting activities, is proposed. Two kinds of time longitudinal analysis method are evaluated using a large amount of actual tweets.

The purpose of this paper is to determine the effectiveness of using fully homomorphic encryption (FHE) to preserve the privacy of biometric data in an authentication system. Biometrics offers higher accuracy for personal recognition than traditional methods because of its properties. Biometric data are permanently linked with an individual and cannot be revoked or cancelled, especially when biometric data are compromised, leading to privacy issues.

By reviewing current approaches, FHE is considered as a promising solution for the privacy issue because of its ability to perform computations in the encrypted domain. The authors studied the effectiveness of FHE in biometric authentication systems. In doing so, the authors undertake the study by implementing a protocol for biometric authentication system using iris.

The security analysis of the implementation scheme demonstrates the effectiveness of FHE to protect the privacy of biometric data, as unlimited operations can be performed in the encrypted domain, and the FHE secret key is not shared with any other party during the authentication protocol.

The use of malicious model in the design of the authentication protocol to improve the privacy, packing methods and use of low-level programming language to enhance performance of the system needs to be further investigated.

The main contributions of this paper are the implementation of a privacy-preserving iris biometric authentication protocol adapted to lattice-based FHE and a sound security analysis of authentication and privacy.

The purpose of this paper is to address the challenge of opinion mining in text documents to perform further analysis such as community detection and consistency control. More specifically, we aim to identify and extract opinions from natural language documents and to represent them in a structured manner to identify communities of opinion holders based on their common opinions. Another goal is to rapidly identify similar or contradictory opinions on a target issued by different holders.

For the opinion extraction problem we opted for a supervised approach focusing on the feature selection problem to improve our classification results. On the community detection problem, we rely on the Infomap community detection algorithm and the multi-scale community detection framework used on a graph representation based on the available opinions and social data.

The classification performance in terms of precision and recall was significantly improved by adding a set of “meta-features” based on grouping rules of certain part of speech (POS) instead of the actual words. Concerning the evaluation of the community detection feature, we have used two quality metrics: the network modularity and the normalized mutual information (NMI). We evaluated seven one-target similarity functions and ten multi-target aggregation functions and concluded that linear functions perform poorly for data sets with multiple targets, while functions that calculate the average similarity have greater resilience to noise.

Although our solution relies on existing approaches, we managed to adapt and integrate them in an efficient manner. Based on the initial experimental results obtained, we managed to integrate original enhancements to improve the performance of the obtained results.

The purpose of this paper is to propose a method of calculating the sentiment value of a tweet based on the emoticon role.

Classification of emoticon roles as four types showing “emphasis”, “assuagement”, “conversion” and “addition”, with roles determined based on the respective relations to sentiment of sentences and emoticons.

Clustering of users of four types based on emoticon sentiment.

Formalization, using regression analysis, of the relation of sentiment between sentences and emoticons in all roles.

This paper aims to propose a multi-label method that estimates appropriate aspects against unknown tweets using the two-phase estimation method. Many Twitter users share daily events and opinions. Some beneficial comments are posted on such real-life aspects as eating, traffic, weather and so on. Such posts as “The train is not coming” are categorized in the Traffic aspect. Such tweets as “The train is delayed by heavy rain” are categorized in both the Traffic and Weather aspects.

The proposed method consists of two phases. In the first, many topics are extracted from a sea of tweets using Latent Dirichlet Allocation (LDA). In the second, associations among many topics and fewer aspects are built using a small set of labeled tweets. The aspect scores for tweets were calculated using associations based on the extracted terms. Appropriate aspects are labeled for unknown tweets by averaging the aspect scores.

Using a large amount of actual tweets, the sophisticated experimental evaluations demonstrate the high efficiency of the proposed multi-label classification method. It is confirmed that high F-measure aspects are strongly associated with topics that have high relevance. Low F-measure aspects are associated with topics that are connected to many other aspects.

The proposed method features two-phase semi-supervised learning. Many topics are extracted using an unsupervised learning model called LDA. Associations among many topics and fewer aspects are built using labeled tweets.