Research Interests:
Research Interests:
While public conversations in Twitter have gained increasing interest in the marketing sector, relatively very little data-mining research have been conducted in this area. In this paper, we empirically evaluate whether employing reply... more
While public conversations in Twitter have gained increasing interest in the marketing sector, relatively very little data-mining research have been conducted in this area. In this paper, we empirically evaluate whether employing reply links in public conversations can enhance the product feature extraction from tweets. We introduce a conversation-based method that considers a conversation as a reply tree and employs anaphora resolution in a backtracking mechanism to effectively extract the product features involved in the messages. We also develop a conversation filtering process based on a set of filtering measures including content relevance and social metrics. We conducted our experiments using a manually annotated Twitter corpus involving smartphones and other electronics products. The experimental results show the effectiveness of our proposed method.
Research Interests:
In this paper, we aim at modeling the user profile containing timely relevant information extracted from his interactions with search engines. We considered a time-sensitive user profile that provides relevant and fresh information... more
In this paper, we aim at modeling the user profile containing timely relevant information extracted from his interactions with search engines. We considered a time-sensitive user profile that provides relevant and fresh information inferred from his submitted queries, reformulated queries and clicked results. We used a unique profile that integrates current and recurrent interactions within a session giving more importance to recent interactions without ignoring the old ones. We conducted experiments using the 2013 TREC Session track and the ClueWeb12 collection that showed the effectiveness of our approach compared to state-of-the-art ones.
Research Interests:
In this paper we focus on the identification of the author of a written text. We present a new hybrid method that combines a set of stylistic and statistical features in a machine learning process. We tested the effectiveness of the... more
In this paper we focus on the identification of the author of a written text. We present a new hybrid method that combines a set of stylistic and statistical features in a machine learning process. We tested the effectiveness of the linguistic and statistical features combined with the inter-textual distance "Delta" on the PAN’@CLEF’2015 English corpus and we obtained 0.59 as c@1 precision.
Research Interests:
In a ubiquitous scenario, people are typically confronted with context evolution and changing influences. This may create new needs and may condition the user perception of what is relevant information. Over the years, different... more
In a ubiquitous scenario, people are typically confronted with context evolution and changing influences. This may create new needs and may condition the user perception of what is relevant information. Over the years, different approaches have been proposed to design personalized Recommender Systems (RS), but state-of-the-art approaches mostly assume a fixed representation of a user profile; the dynamicity of the user's interests (and the way of expressing them) while interacting with the environment is not considered. Aim of this work is to predict a user's preferences in the tourism domain, to provide personalized and context-aware recommendations. Therefore, we define a user profile model which expresses in a formal way the user's opinions with respect to a particular entity. In particular, the proposed approach formally models the user generated content (UGC) connected to a group of reviews (written by expert users) for each entity, and compares it with a (positive and negative) statistical language model representing the target user profile associated with that entity. The effectiveness of the approach is illustrated on a real-case scenario.
Research Interests:
With the huge amount of daily generated social networks posts, reviews, ratings, recommendations and other forms of online expressions, the web 2.0 has turned into a crucial opinion rich resource. Since others’ opinions seem to be... more
With the huge amount of daily generated social networks posts, reviews, ratings, recommendations and other forms of online expressions, the web 2.0 has turned into a crucial opinion rich resource. Since others’ opinions seem to be determinant when making a decision both on individual and organizational level, several researches are currently looking to the sentiment analysis.
Research Interests:
Research Interests:
Community Question Answering (cQA) continues to gain momentum owing to the unceasing rise of user-generated content that dominates the web. CQA are platforms that enable people with different backgrounds to share knowledge by freely... more
Community Question Answering (cQA) continues to gain momentum owing to the unceasing rise of user-generated content that dominates the web. CQA are platforms that enable people with different backgrounds to share knowledge by freely asking and answering each other. In this paper, we focus on question retrieval which is deemed to be a key task in cQA. It aims at finding similar archived questions given a new query, assuming that the answers to the similar questions should also answer the new one. This is known to be a challenging task due to the ver-boseness in natural language and the word mismatch between the questions. Most traditional methods measure the similarity between questions based on the bag-of-words (BOWs) representation capturing no semantics between words. In this paper , we rely on word representation to capture the words semantic information in language vector space. Questions are then ranked using cosine similarity based on the vector-based word representation for e...
Research Interests:
L’etude de l’influence sur Twitter est un sujet de recherche intense, certains utilisateurs revelent plus de capacite pour influencer d’autres personnes. Nous proposons une nouvelle approche pour une evaluation de l’influence polarisee... more
L’etude de l’influence sur Twitter est un sujet de recherche intense, certains utilisateurs revelent plus de capacite pour influencer d’autres personnes. Nous proposons une nouvelle approche pour une evaluation de l’influence polarisee dans les reseaux multirelationnels tels que Twitter. Nous prenons en compte le contenu des tweets pour determiner leur polarite en utilisant l’algorithme des forets d’arbres decisionnels. Puis, nous fusionnons, au moyen des fonctions de croyance, les informations provenant des relations (retweet, mention ou repond, etc.) pour obtenir un degre d’influence pour chaque utilisateur. Nous experimentons notre methode sur les donnees collectees lors des elections europeennes de 2014. Les resultats montrent que notre modele est suffisamment flexible pour repondre aux besoins des specialistes en sciences sociales et que l’utilisation de la theorie des fonctions de croyances est efficace pour traiter des relationsmultiples.
Research Interests:
Research Interests:
Research Interests:
Question Answering is most likely one of the toughest tasks in the field of Natural Language Processing. It aims at directly returning accurate and short answers to questions asked by users in human language over a huge collection of... more
Question Answering is most likely one of the toughest tasks in the field of Natural Language Processing. It aims at directly returning accurate and short answers to questions asked by users in human language over a huge collection of documents or database. Recently, the continuously exponential rise of digital information has imposed the need for more direct access to relevant answers. Thus, question answering has been the subject of a widespread attention and has been extensively explored over the last few years. Retrieving passages remains a crucial but also a challenging task in question answering. Although there has been an abundance of work on this task, this latter still implies non-trivial endeavor. In this paper, we propose an ad-hoc passage retrieval approach for Question Answering using n-grams. This approach relies on a new measure of similarity between a passage and a question for the extraction and ranking of the different passages based on n-gram overlapping. More conc...
Research Interests:
Research Interests:
The growing complexity of the Twitter micro-blogging service in terms of size, number of users, and variety of bloggers relationships have generated a big data which requires innovative approaches in order to analyse, extract and detect... more
The growing complexity of the Twitter micro-blogging service in terms of size, number of users, and variety of bloggers relationships have generated a big data which requires innovative approaches in order to analyse, extract and detect non-obvious and popular events. Under such a circumstance, we aim, in this paper, to use big data analytics within twitter to allow real time event detection. These challenges present a big opportunity for Natural Language Processing (NLP) and Information Extraction (IE) technology to enable new large-scale data-analysis applications. Taking to account all the difficulties, this paper proposes a new metric to improve the results of the searches in microblogs. It combines content relevance, tweet relevance and author relevance, and develops a Natural Language Processing method for extracting temporal information of events from posts more specifically tweets. Our approach is based on a methodology of temporal markers classes and on a contextual exploration method. To evaluate our model, we built a knowledge management system. Actually, we used a collection of 10 thousand of tweets talking about the current events in 2014 and 2015.
Research Interests:
ABSTRACT In this paper, we are interested in aggregated search in structured XML documents. We present a structured information retrieval model based on the Bayesian networks theory. Query-terms and terms-elements relations are modeled... more
ABSTRACT In this paper, we are interested in aggregated search in structured XML documents. We present a structured information retrieval model based on the Bayesian networks theory. Query-terms and terms-elements relations are modeled through probability. In this model, the user's query starts a propagation process to recover the XML elements. Thus, instead of retrieving a whole document or a list of disjoint elements that are likely to answer partially the query, we attempt to built a virtual document that aggregates a set of elements, that are relevant all together. We evaluated our approach using the INEX 2009 collection and presented some empirical results for evaluating the impact of the aggregation approach.
Research Interests:
Research Interests:
ABSTRACT The machine learning methods like support vectors machines, hidden Markov model and conditional random fields are the most used methods for implementing natural language processing systems. In this paper, we propose a machine... more
ABSTRACT The machine learning methods like support vectors machines, hidden Markov model and conditional random fields are the most used methods for implementing natural language processing systems. In this paper, we propose a machine learning approach that can be used for sequential labeling tasks like biological event extraction. Our biological event extraction approach uses Support Vector Machines (SVM) and a composite kernel function to identify triggers and to assign the corresponding arguments. Also, we use a number of features based on both syntactic and contextual information which were automatically learned from the training data.
Research Interests:
The use of text similarity plays an important role in many applications in Computational Linguistics, such as Text Classification and Information Extraction and Retrieval. Besides, there are several tasks that require computing the... more
The use of text similarity plays an important role in many applications in Computational Linguistics, such as Text Classification and Information Extraction and Retrieval. Besides, there are several tasks that require computing the similarity between two short segments of text. In this work, we propose a sentence similarity computing approach that takes account of the semantic and the syntactic information contained in the sentences. The proposed method can be applied in a variety of applications to mention, text knowledge representation and discovery. Experiments on a set of sentence pairs show that our approach presents a similarity measure that illustrates a considerable correlation to human judgment.
Research Interests:
Research Interests:
Research Interests:
After the beginning of the extension of current Web towards the semantics, the annotation starts to take a significant role, since it takes part to give the semantic aspect to the different types of documents. With the proliferation of... more
After the beginning of the extension of current Web towards the semantics, the annotation starts to take a significant role, since it takes part to give the semantic aspect to the different types of documents. With the proliferation of news articles from thousands of different sources now available on the Web, summarization of such information is becoming increasingly important. We will define a methodological approach to extract the events from the news articles and to annotate them according to the principal events which they contain. Considering the large number of news source (for examples, BBC, Reuters, CNN…), every day, thousands of articles are produced in the entire world concerning a given event. This is why we have to think to automate the process of annotation of such articles.
Résumé. L’utilisation des documents pédagogiques, disponibles sur le web, devient de plus en plus large tant pour l’enseignant qui a besoin de préparer son support de cours que pour l’étudiant qui désire, par exemple, s’autoformer. La... more
Résumé. L’utilisation des documents pédagogiques, disponibles sur le web, devient de plus en plus large tant pour l’enseignant qui a besoin de préparer son support de cours que pour l’étudiant qui désire, par exemple, s’autoformer. La description d’un document pédagogique, en l’alimentant par des métadonnées, s’avère une solution qui confère une valeur ajoutée au document afin d’expliciter des informations placées dans ce document. Dans cette optique, nous proposons une méthode d’annotation de documents pédagogiques selon différents points de vue, qui est basée sur l’analyse sémantique des éléments discursifs du texte.
Research Interests:
A new challenge is added to the Natural Language Processing Community; how to analyze the new documents forms resulting from the Web 2.0? We are interested in a particular kind of information which is events. Thus, we propose a generic... more
A new challenge is added to the Natural Language Processing Community; how to analyze the new documents forms resulting from the Web 2.0? We are interested in a particular kind of information which is events. Thus, we propose a generic approach to extract and analyze events from text. We propose an event extraction algorithm with a polynomial complexity O(n). This algorithm is based on developed semantic map of events. We validate the first component of our approach by the development of the "EventEC" system.
Research Interests:
In this paper we focus on the identification of the author of a written text. We present a new hybrid method that combines a set of stylistic and statistical features in a machine learning process. We tested the effectiveness of the... more
In this paper we focus on the identification of the author of a written text. We present a new hybrid method that combines a set of stylistic and statistical features in a machine learning process. We tested the effectiveness of the linguistic and statistical features combined with the inter-textual distance "Delta" on the PAN’@CLEF’2015 English corpus and we obtained 0.59 as c@1 precision.
This paper focuses on question retrieval which is a crucial and tricky task in Community Question Answering (cQA). Question retrieval aims at finding historical questions that are semantically equivalent to the queried ones, assuming that... more
This paper focuses on question retrieval which is a crucial and tricky task in Community Question Answering (cQA). Question retrieval aims at finding historical questions that are semantically equivalent to the queried ones, assuming that the answers to the similar questions should also answer the new ones. The major challenges are the lexical gap problem as well as the verboseness in natural language. Most existing methods measure the similarity between questions based on the bag-of-words (BOWs) representation capturing no semantics between words. In this paper, we rely on word embeddings and TF-IDF for a meaningful vector representation of the questions. The similarity between questions is measured using cosine similarity based on their vector-based word representations. Experiments carried out on a real world data set from Yahoo! Answers show that our method is competetive.
Research Interests:
When looking for recently published scientific papers, a researcher usually focuses on the topics related to her/his scientific interests. The task of a recommender system is to provide a list of unseen papers that match these topics. The... more
When looking for recently published scientific papers, a researcher usually focuses on the topics related to her/his scientific interests. The task of a recommender system is to provide a list of unseen papers that match these topics. The core idea of this paper is to leverage the latent topics of interest in the publications of the researchers, and to take advantage of the social structure of the researchers (relations among researchers in the same field) as reliable sources of knowledge to improve the recommendation effectiveness. In particular, we introduce a hybrid approach to the task of scientific papers recommendation, which combines content analysis based on probabilistic topic modeling and ideas from collaborative filtering based on a relevance-based language model. We conducted an experimental study on DBLP, which demonstrates that our approach is promising.
Research Interests:
RÉSUMÉ. L’influence sur Twitter est devenue un sujet de recherche important. Certains utilisateurs révèlent plus de capacité que d’autres pour influencer les personnes avec lesquelles ils sont connectés. Ainsi, trouver les utilisateurs... more
RÉSUMÉ. L’influence sur Twitter est devenue un sujet de recherche important. Certains utilisateurs révèlent plus de capacité que d’autres pour influencer les personnes avec lesquelles ils sont connectés. Ainsi, trouver les utilisateurs les plus influents peut permettre une diffusion efficace de l’information à grande échelle, action très utile dans le marketing ou les campagnes politiques. Dans cet article, nous proposons une nouvelle approche pour l’évaluation de l’influence dans les réseaux multi-relationnels tels que Twitter. Notre méthode est basée sur la règle de combinaison conjonctive de la théorie des fonctions de croyance qui permet de fusionner différents types de relations. Nous expérimentons notre méthode sur des données Twitter collectées lors des élections européennes de 2014 et déterminons les candidats les plus influents.
Research Interests:
The Context Aware Recommender Systems aim to combine a set of technologies and knowledge about the user context not only in order to deliver the most appropriate information to the user need at just the right time. It is called Proactive... more
The Context Aware Recommender Systems aim to combine a set of technologies and knowledge about the user context not only in order to deliver the most appropriate information to the user need at just the right time. It is called Proactive Recommendation. In this paper, we present a project funded by the European Support Programme for Research and Innovation1 (MOBIDOC). This project aims to bring the “just-in-time information” and the contextual dimension to geo-based systems developed by the company TUNAV2 to provide users with customized products tailored to their own needs and preferences. The main idea we plan to explore is the integration of the user profile (preferences, needs, etc) collected from the Social Networks and his/her navigation tendency in the products offered by TUNAV, in order to recommend to the user synthesized and relevant information without having to wait for a query.In our work, Social Networks will play a double-edged role as a foundation for user context mo...
Research Interests:
Les systemes de recommandation contextuelle visent a combiner un ensemble de technologies et de connaissances sur le contexte de l’utilisateur pour lui fournir une information pertinente au moment ou il en a le plus besoin, c’est ce qu’on... more
Les systemes de recommandation contextuelle visent a combiner un ensemble de technologies et de connaissances sur le contexte de l’utilisateur pour lui fournir une information pertinente au moment ou il en a le plus besoin, c’est ce qu’on appelle la recommandation proactive. Dans cet article nous proposons une approche de recommandation contextuelle et proactive dans un environnement mobile qui apprend implicitement les preferences de l’utilisateur. Nous avons evalue notre approche dans le cadre de la tâche “Contextual Suggestion Track” de TREC 2014. Les resultats que nous avons obtenus sont prometteurs.
Research Interests:
With the data volume that does not stop growing and the multitude of sources that led to diversity of structures, the classic tools of data management became unsuitable for processing and unable to offer effective tools for information... more
With the data volume that does not stop growing and the multitude of sources that led to diversity of structures, the classic tools of data management became unsuitable for processing and unable to offer effective tools for information retrieval and knowledge management. Thereby, a major challenge has become how to deal with the explosion of data to transform it into new useful and interesting knowledge. Despite the rapid development and change of the databases world, this data management systems diversity presents a difficulty in choosing the best solution to analyze, interpret and manage data according to the user’s needs while preserving data availability. Hence, the imposition of the Big Data in our techno-logical landscape offers new solutions for data processing. In this work, we aim to present a brief of the current buzz research field called Big Data. Then, we provide a broad comparison of two data management technologies.
Research Interests:
In this paper, we deal with the author identification issues of the document whose origin is unknown. To overcome these problems, we propose a new hybrid approach combining the statistical and stylistic analysis. Our introduced method is... more
In this paper, we deal with the author identification issues of the document whose origin is unknown. To overcome these problems, we propose a new hybrid approach combining the statistical and stylistic analysis. Our introduced method is based on determining the lexical and syntactic features of the written text in order to identify the author of the document. These features are explored to build a machine learning process. We obtained promising results by relying on PAN@CLEF2014 English literature corpus. The experimental results are comparable to those obtained by the best state of the art methods.
Research Interests:
In this paper, we propose a new method for opinion target identification based on twitter conversations rather than simple individual tweets. We employ conversation interactions to effectively extract the different target features using a... more
In this paper, we propose a new method for opinion target identification based on twitter conversations rather than simple individual tweets. We employ conversation interactions to effectively extract the different target features using a product review corpus involving smart-phones and other electronics products. Experimental evaluations show that our proposed method is efficient and contributes to improving system performance.
Research Interests:
Un systeme de recherche d’information performant doit satisfaire les differents types de besoins des utilisateurs visant une variete de categories de requetes. Ces categories comprennent les requetes sensibles au temps ou le contenu... more
Un systeme de recherche d’information performant doit satisfaire les differents types de besoins des utilisateurs visant une variete de categories de requetes. Ces categories comprennent les requetes sensibles au temps ou le contenu recent est l’exigence principale de l’utilisateur. Cependant, l’utilisation des caracteristiques temporelles des documents pour mesurer leur fraicheur reste une tâche difficile etant donne que ces caracteristiques ne sont pas representees avec precision dans les documents recents. Dans cet article, nous proposons un modele de langue qui estime la pertinence et la fraicheur des documents vis-a-vis des requetes sensibles au temps reel. Notre approche modelise la fraicheur en exploitant des sources d’information fraiches, plus precisement des termes extraits a partir des tweets recents et thematiquement pertinents par rapport a la requete de l’utilisateur. Nos experimentations montrent que l’extraction des termes frais a partir de Twitter ameliore les resul...
Research Interests:
Semantic similarity between words is fundamental to various fields such as Cognitive Science, Artificial Intelligence, Natural Language Processing and Information Retrieval. According to Baeza-Yates and Neto [2] an Information Retrieval... more
Semantic similarity between words is fundamental to various fields such as Cognitive Science, Artificial Intelligence, Natural Language Processing and Information Retrieval. According to Baeza-Yates and Neto [2] an Information Retrieval system “should provide the user with easy access to the information in which he is interested”. Therefore, in this domain, relying on a robust semantic similarity measure is crucial for automatic query suggestion and expansion process. In this same context, we propose a method that uses on one hand, an online English dictionary provided by the Semantic Atlas project of the French National Centre for Scientific Research (CNRS) and on the other hand, a page counts based metric returned by a social website.
With the data volume that does not stop growing and the multitude of sources that led to diversity of structures, data processing needs are changing. Although, relational DBMSs remain the main data management technology for processing... more
With the data volume that does not stop growing and the multitude of sources that led to diversity of structures, data processing needs are changing. Although, relational DBMSs remain the main data management technology for processing structured data, but faced with the massive growth in the volume of data, despite their evolution, relational databases, which have been proven for over 40 years, have reached their limits. Several organizations and researchers turned to MapReduce framework that has found great success in analyzing and processing large amounts of data on large clusters. In this paper, we will discuss MapReduce and Relational Database Management Systems as competing paradigms, and then as completing paradigms where we propose an integration approach to optimize OLAP queries process.
Research Interests:
In this paper, we focus on the problem of question retrieval in community Question Answering (cQA) which aims to retrieve from the community archives the previous questions that are semantically equivalent to the new queries. The major... more
In this paper, we focus on the problem of question retrieval in community Question Answering (cQA) which aims to retrieve from the community archives the previous questions that are semantically equivalent to the new queries. The major challenges in this crucial task are the shortness of the questions as well as the word mismatch problem as users can formulate the same query using different wording. While numerous attempts have been made to address this problem, most existing methods relied on supervised models which significantly depend on large training data sets and manual feature engineering. Such methods are mostly constrained by their specificities that put aside the word order and ignore syntactic and semantic relationships. In this work, we rely on Neural Networks (NNs) which can learn rich dense representations of text data and enable the prediction of the textual similarity between the community questions. We propose a deep learning approach based on a Siamese architecture...
Plusieurs utilisateurs ont souvent besoin d’informations pédagogiques pour les intégrer dans leurs ressources pédagogiques, ou pour les utiliser dans un processus d’apprentissage. Une indexation de ces informations s’avère donc utile en... more
Plusieurs utilisateurs ont souvent besoin d’informations pédagogiques pour les intégrer dans leurs ressources pédagogiques, ou pour les utiliser dans un processus d’apprentissage. Une indexation de ces informations s’avère donc utile en vue d’une extraction des informations pédagogiques pertinentes en réponse à une requête utilisateur. La plupart des systèmes d’extraction d’informations pédagogiques existants proposent une indexation basée sur une annotation manuelle ou semi-automatique des informations pédagogiques, tâche qui n’est pas préférée par les utilisateurs. Dans cet article, nous proposons une approche d’indexation d’objets pédagogiques (Définition, Exemple, Exercice, etc.) basée sur une annotation sémantique par Exploration Contextuelle des documents. L’index généré servira à une extraction des objets pertinents répondant à une requête utilisateur sémantique. Nous procédons, ensuite, à un classement des objets extraits selon leur pertinence en utilisant l’algorithme Rocch...
Recently, there has been a burgeoning interest in keyword search in relational databases owing to its ease of use. Although extensive research has been lately done within this context, most of this research not only requires a prior... more
Recently, there has been a burgeoning interest in keyword search in relational databases owing to its ease of use. Although extensive research has been lately done within this context, most of this research not only requires a prior access to data which severely restricts their applicability if this condition is not verified, but also returns very generic answers. However, providing users with personalized answers has become more than ever necessary due to the overabundance of data which can be annoying for the user. The challenge to return personalized and relevant answers that satisfy users’ information needs remains. Inspired by the successful application of the collaborative filtering technique in recommender systems, we propose a novel keyword-based approach to provide users with personalized results based on the hypothesis that only information on the database schema is available.
This paper presents an approach for author profiling of an unknown users from their texts produced in social media. In particular, we address the identification of two profile dimensions: gender and language variety, of Arabic twitter... more
This paper presents an approach for author profiling of an unknown users from their texts produced in social media. In particular, we address the identification of two profile dimensions: gender and language variety, of Arabic twitter users based on their tweets. Our approach focused on applying metaclassification technique on features extracted from tweets body. We explored two main sets of features which are character and word n-grams. The proposed approach allowed us to reach promising results for both language variety and gender identification
In the current era, microblogging services like Twitter, gives people the ability to communicate, interact, collaborate with each other, reply to messages from others and create conversations. These services can be seen as very large... more
In the current era, microblogging services like Twitter, gives people the ability to communicate, interact, collaborate with each other, reply to messages from others and create conversations. These services can be seen as very large information repository containing millions of text messages usually organized into complex networks involving users interacting with each other at specific times. Several works have proposed tools for tweets search focused only to retrieve relevant tweets. Therefore, users are unable to explore the results or retrieve more relevant tweets based on the content, and may get lost or become frustrated by the information overload. In this paper, we propose a new method to retrieve conversation on microblog-ging sites particularly Twitter. It's based on content analysis and content enrichment. The goal of our method is to present a more informative result compared to conventional search engine. The proposed method has been implemented and evaluated by com...
Just-In-Time Recommender Systems involve all systems able to provide recommendations tailored to the preferences and needs of users in order to help them access useful and interesting resources within a large data space. The user does not... more
Just-In-Time Recommender Systems involve all systems able to provide recommendations tailored to the preferences and needs of users in order to help them access useful and interesting resources within a large data space. The user does not need to formulate a query, this latter is implicit and corresponds to the resources that match the user's interests at the right time. In this paper, we propose a proactive context-aware recommendation approach for mobile devices that covers many domains. It aims at recommending relevant items that match users' personal interests at the right time without waiting for users to initiate any interaction.
Nowadays, sentiment analysis methods become more and more popular especially with the proliferation of social media platform users number. In the same context, this paper presents a sentiment analysis approach which can faithfully... more
Nowadays, sentiment analysis methods become more and more popular especially with the proliferation of social media platform users number. In the same context, this paper presents a sentiment analysis approach which can faithfully translate the sentimental orientation of Arabic Twitter posts, based on a novel data representation and machine learning techniques. The proposed approach applied a wide range of features: lexical, surface-form, syntactic, etc. We also made use of lexicon features inferred from two Arabic sentiment words lexicons. To build our supervised sentiment analysis system, we use several standard classification methods (Support Vector Machines, K-Nearest Neighbour, Naïve Bayes, Decision Trees, Random Forest) known by their effectiveness over such classification issues. In our study, Support Vector Machines classifier outperforms other supervised algorithms in Arabic Twitter sentiment analysis. Via an ablation experiments, we show the positive impact of lexicon based features on providing higher prediction performance.
Research Interests:
Research Interests:
Research Interests:
Research Interests:
The conversational element of Twitter has recently become of particular interest to the marketing community. However, most studies on mining product features through Twitter, have so far employed simple individual tweets rather than... more
The conversational element of Twitter has recently become of particular interest to the marketing community. However, most studies on mining product features through Twitter, have so far employed simple individual tweets rather than considering the whole conversations. In this paper, we empirically evaluate whether employing user interactions in public conversations can improve the product feature extraction from tweets. We propose a conversation-based method which considers a conversation as a reply tree and employs reply links, to effectively extract the product features involved in the messages. We also develop a conversation filtering process which combines scores measured from different aspects including content relevance and social aspects. We conducted our experiments using a manually annotated Twitter corpus involving smartphones and other electronics products. The experimental results show the effectiveness of our proposed method.
Research Interests:
This paper puts forward a new recommendation algorithm based on semantic analysis as well as new measurements. Like Facebook, Social network is considered as one of the most well-prominent Web 2.0 applications and relevant services... more
This paper puts forward a new recommendation algorithm based on semantic analysis as well as new measurements. Like Facebook, Social network is considered as one of the most well-prominent Web 2.0 applications and relevant services elaborating into functional ways for sharing opinions. Thereupon, social network web sites have since become valuable data sources for opinion mining. This paper proposes to introduce an external resource a sentiment from comments posted by users in order to anticipate recommendation and also to lessen the cold-start problem. The originality of the suggested approach means that posts are not merely characterized by an opinion score, but receive an opinion grade notion in the post instead. In general, the authors' approach has been implemented with Java and Lenskit framework. The study resulted in two real data sets, namely MovieLens and TripAdvisor, in which the authors have shown positive results. They compared their algorithm to SVD and Slope One al...