Abstract Latent Dirichlet Allocation (LDA) is the technique most commonly used in topic modeling methods, but it requires the number of topics generated by LDA to be specified for topic recognition modeling. Except the main iterative methods based on perplexity and nonparametric methods, recent research has no simple way to select the optimal number of topics in the model. Aiming... Latent Dirichlet Allocation (LDA) is the technique most commonly used in topic modeling methods, but it requires the number of topics generated by LDA to be specified for topic recognition modeling. Except the main iterative methods based on perplexity and nonparametric methods, recent research has no simple way to select the optimal number of topics in the model. Aiming at appropriately determining the number of topics and then optimizing the LDA topic model, this paper proposes a non-iterative method for automatically determining the number of topics. The clustering method is based on fast seeking and locating density peaks. This method transforms the traditional topic cluster number selection problem into clustering problem and thus can be used to optimize the topic recognition model for news texts. It does not need iterative optimization and can simplify model development. This method uses Word2Vec for word embedding on corpus text to explore the superior performance of word-related relationships and to express the implicit semantic relationship between topic corpora. Then, using a clustering algorithm that quickly searches for and finds the cluster peaks; the word vectors after word embedding are clustered to obtain the number of word vector clusters after word embedding. The number of clusters is used as the number of topics in the text. Finally, the experimental results show that the proposed method enjoys better precision and F1 value than the perplexity based method, and is suitable for the identification of the number of topics in corpora in different sizes. This method can effectively find the appropriate number of topics from the news text dataset and improve the accuracy of the LDA theme model. Read More Read Less

Abstract Today, enterprises use a variety of applications to manage day-by-day business activities using a large centralized database. Since a huge amount of data stored in this centralized database produced by the daily use of several systems, it is important to integrate decision- making tools to analyse and interpret these business data. For this purpose, Data Mining is a powerful technology that promote information and knowledge extraction from large... Today, enterprises use a variety of applications to manage day-by-day business activities using a large centralized database. Since a huge amount of data stored in this centralized database produced by the daily use of several systems, it is important to integrate decision- making tools to analyse and interpret these business data. For this purpose, Data Mining is a powerful technology that promote information and knowledge extraction from large databases. In this paper, we present an agent-based approach for extracting business association rules from centralized database systems. This approach combine paradigm of multi-agent system and the association rules as a data mining technique to build anefficient model. It is relying on the intelligent partitioning of data to make the execution of business association rules in a parallel and distributed way from a large centralized database. To validate our approach, we applied it during the realization of a real case study on ERP database at the National company of Well Services (ENSP)using JADE platform with machine learning WEKA toolbox for association rules mining. The developed system has been compared with the classic association rules algorithms and has proved it is more efficient and more scalable. The main objective of our work is to improve and accelerate the process of extracting association rules by business through centralized database systems. As a result, the decision process of these systems becomes more improved. Read More Read Less

Abstract This paper presents an approach based on supervised machine learning methods to discriminate between positive, negative and neutral Arabic reviews in online newswire. The corpus is labeled for subjectivity and sentiment analysis (SSA) at the sentence-level. The model uses both count and TF-IDF representations and apply six machine learning algorithms; Multinomial Naïve Bayes, Support Vector Machines (SVM), Random Forest, Logistic Regression, Multi-layer perceptron and k-nearest neighbors using uni-grams, bi-grams... This paper presents an approach based on supervised machine learning methods to discriminate between positive, negative and neutral Arabic reviews in online newswire. The corpus is labeled for subjectivity and sentiment analysis (SSA) at the sentence-level. The model uses both count and TF-IDF representations and apply six machine learning algorithms; Multinomial Naïve Bayes, Support Vector Machines (SVM), Random Forest, Logistic Regression, Multi-layer perceptron and k-nearest neighbors using uni-grams, bi-grams features. With the goal of extracting users' sentiment from written text. Experimental results showed that n-gram features could substantially improve performance; and showed that the Multinomial Naive Bayes approach is the most accurate in predicting topic polarity. Best results were achieved using count vectors trained by combination of word-based uni-gramsand bi-grams with an overall accuracy of 85.57% over two classes and 65.64% over three classes. Read More Read Less

Abstract State-of-the-art sentiment analysis systems rely on a sentiment lexicon, which is the most essential feature that drives their performance. This resource is indispensable for, and greatly contributes to, sentiment analysis tasks. This is evident in the emergence of a large volume of research devoted to the development of automated sentiment lexicon generation algorithms. The task of tagging subjective words with... State-of-the-art sentiment analysis systems rely on a sentiment lexicon, which is the most essential feature that drives their performance. This resource is indispensable for, and greatly contributes to, sentiment analysis tasks. This is evident in the emergence of a large volume of research devoted to the development of automated sentiment lexicon generation algorithms. The task of tagging subjective words with a semantic orientation comprises two core approaches: dictionary based and corpus-based. The former involves making use of an online dictionary to tag words, while the latter relies on co-occurrence statistics or syntactic patterns embedded in text corpora. The end result is a linguistic resource comprising a priori information about words, across the semantic dimension of sentiment. This paper provides a survey on the most prominent research works that utilize corpus-based techniques for sentiment lexicon generation. We also conduct a comparative analysis on the performance of state-of-the-art algorithms proposed for this task, and shed light on the current progress and challenges in this area. Read More Read Less