An artificial neural network based approach for sentiment analysis of opinionated text
23 Oct 2012-pp 37-42
TL;DR: A sentiment classification model using back-propagation artificial neural network (BPANN) is proposed that combines the strength of BPANN in classification accuracy with utilizing intrinsic domain knowledge available in the sentiment lexicons.
Abstract: The Internet and Web 2.0 social media have emerged as an important medium for expressing sentiments, opinions, evaluations, and reviews. Sentiment analysis or opinion mining is becoming an open research domain due to the abundance of discussion forums, Weblogs, e-commerce portals, social networking and content sharing sites where people tend to express their opinions. Sentiment Analysis involves classifying text documents based on the opinion expressed being positive or negative about a given topic. This paper proposes a sentiment classification model using back-propagation artificial neural network (BPANN). Information Gain and three popular sentiment lexicons are used to extract sentiment representing features that are then used to train and test the BPANN. This novel approach combines the strength of BPANN in classification accuracy with utilizing intrinsic domain knowledge available in the sentiment lexicons. The results obtained on the movie-review corpora have shown that the proposed approach has been able to reduce dimensionality, while producing accurate sentiment based classification of text.
Citations
More filters
TL;DR: An encoding-based Wave2vec time series classifier model, which combines signal-processing and deep learning-based natural language processing techniques, is provided, which facilitates intuitive and easy recognition, and identification of influential patterns.
Abstract: Interest in research involving health-medical information analysis based on artificial intelligence, especially for deep learning techniques, has recently been increasing. Most of the research in this field has been focused on searching for new knowledge for predicting and diagnosing disease by revealing the relation between disease and various information features of data. These features are extracted by analyzing various clinical pathology data, such as EHR (electronic health records), and academic literature using the techniques of data analysis, natural language processing, etc. However, still needed are more research and interest in applying the latest advanced artificial intelligence-based data analysis technique to bio-signal data, which are continuous physiological records, such as EEG (electroencephalography) and ECG (electrocardiogram). Unlike the other types of data, applying deep learning to bio-signal data, which is in the form of time series of real numbers, has many issues that need to be resolved in preprocessing, learning, and analysis. Such issues include leaving feature selection, learning parts that are black boxes, difficulties in recognizing and identifying effective features, high computational complexities, etc. In this paper, to solve these issues, we provide an encoding-based Wave2vec time series classifier model, which combines signal-processing and deep learning-based natural language processing techniques. To demonstrate its advantages, we provide the results of three experiments conducted with EEG data of the University of California Irvine, which are a real-world benchmark bio-signal dataset. After converting the bio-signals (in the form of waves), which are a real number time series, into a sequence of symbols or a sequence of wavelet patterns that are converted into symbols, through encoding, the proposed model vectorizes the symbols by learning the sequence using deep learning-based natural language processing. The models of each class can be constructed through learning from the vectorized wavelet patterns and training data. The implemented models can be used for prediction and diagnosis of diseases by classifying the new data. The proposed method enhanced data readability and intuition of feature selection and learning processes by converting the time series of real number data into sequences of symbols. In addition, it facilitates intuitive and easy recognition, and identification of influential patterns. Furthermore, real-time large-capacity data analysis is facilitated, which is essential in the development of real-time analysis diagnosis systems, by drastically reducing the complexity of calculation without deterioration of analysis performance by data simplification through the encoding process.
22 citations
Cites methods from "An artificial neural network based ..."
...Document classification, another instance of the sequence classification problem, is performed by using the order of words as a major feature [25], a sentimental analysis is performed by using the sentiment lexicons of words [26], and the pattern-based sequence classification is studied in more general areas [27]....
[...]
01 Jan 2016
TL;DR: The system performs well for recurrent neural network when compared with the system submitted to the shared task as the accuracy of the system had increased and the network seeks to pursue sentiment oriented feature which improves in analyzing the sentiments on tweets.
Abstract: This paper aims at improving the system which is submitted to the shared task on Sentiment Analysis in Indian Languages (SAIL2015) at MIKE 2015. In this work the tweets are classified into three polarity category namely positive, negative and neutral. Twitter data of three languages namely Tamil, Hindi and Bengali are already provided by SAIL 2015 task organizers as we have participated in the contest. Recurrent neural network is used for analyzing the sentiment in the tweets. The system performs well for recurrent neural network when compared with the system submitted to the shared task as the accuracy of the system had increased. This is due to the fact that the recurrent neural network concentrates more on language specific feature. In training, the recurrent neural network tries to learn based on the error that are generated as intermediate output. By this way the network seeks to pursue sentiment oriented feature which improves in analyzing the sentiments on tweets. We have obtained a state accuracy for the proposed system, where we achieved an accuracy of 88%, 72.01% and 65.16% for Tamil, Hindi and Bengali languages respectively for SAIL 2015 dataset.
19 citations
Cites methods from "An artificial neural network based ..."
...In this work along with BPANN, it uses domain knowledge which are available in sentiment lexicon [8]....
[...]
TL;DR: The results suggest that institutional and departmental barriers driven by power asymmetries play a large role in the underreporting of sexual harassment among students—especially those in STEM disciplines.
Abstract: What factors predict the underreporting of sexual harassment in academe? We used logistic regression and sentiment analysis to examine 2,343 reports of sexual harassment involving members of university communities. Results indicate students were 1.6 times likely to not report their experiences when compared to faculty. Respondents in the life and physical sciences were 1.7 times more likely to not report their experiences when compared to respondents in other disciplines. Men represented 90% of the reported perpetrators of sexual harassment. Analysis of respondents' written accounts show variation of overall sentiment based on discipline, student type, and the type of institution attended, particularly with regard to mental health. Our results suggest that institutional and departmental barriers driven by power asymmetries play a large role in the underreporting sexual harassment among students-especially those in STEM disciplines.
19 citations
01 Jan 2018
TL;DR: A new topic model based approach for opinion mining and sentiment analysis of text reviews posted in web forums or social media site which are mostly in unstructured in nature is discussed.
Abstract: This paper discusses a new topic model based approach for opinion mining and sentiment analysis of text reviews posted in web forums or social media site which are mostly in unstructured in nature. In recent years, opinions are exchanged in clouds about any product, person, event or any interested topic. These opinions help in decision making for choosing a product or getting feedback about any topic. Opinion mining and sentiment analysis are related in a sense that opining mining deals with analyzing and summarizing expressed opinions whereas sentiment analysis classifies opinionated text into positive and negative. Aspect extraction is a crucial problem in sentiment analysis. Model proposed in the paper utilizes topic model for aspect extraction and support vector machine learning technique for sentiment classification of textual reviews. The goal is to automate the process of mining attitudes, opinions and hidden emotions from text.
12 citations
Cites methods from "An artificial neural network based ..."
...On reviewing literature, it is found that various supervised machine learning algorithms like Naïve Bayes [12] [13] Support Vector Machines[14] [15] and Neural Networks [16] have been used for opinion mining of text to classify positive and negative sentiment....
[...]
01 Jul 2016
TL;DR: This work proposes an experimental study between their approach using Fuzzy Systems and an execution using Artificial Neural Network to verify which is the most appropriate to solve the problem to estimate the importance of reviews.
Abstract: With the evolution of e-commerce and Online Social Networks, the web information has constantly increased, so the relevance to create methods for automatic knowledge extraction and data mining earned notoriety. Information as opinion evaluation is a point studied by Sentiment Analysis area, which is becoming important nowadays. Be aware of the best reviews is a factor that must be taken into account. Sousa et al. proposed an approach to estimate the degree of importance of reviews about product and services using Fuzzy System, reporting good results. This work proposes an experimental study between their approach using Fuzzy Systems and an execution using Artificial Neural Network to verify which is the most appropriate to solve the problem to estimate the importance of reviews.
12 citations
Cites methods from "An artificial neural network based ..."
...Sharma and Dey [10] proposed a sentiment classification model using backpropagation artificial neural network, using information gain and three popular sentiment lexicons to extract sentiment representing features that are then used to train and test the network....
[...]
References
More filters
23 Jul 2007
TL;DR: This paper proposes to use some linguistic rules to deal with the problem of determining the semantic orientations (positive or negative) of opinions expressed on product features in reviews together with a new opinion aggregation function.
Abstract: Online product reviews are one of the important opinion sources on the Web. This paper studies the problem of determining the semantic orientations (positive or negative) of opinions expressed on product features in reviews. Most existing approaches use a set of opinion words for the purpose. However, the semantic orientations of many words are context dependent. In this paper, we propose to use some linguistic rules to deal with the problem together with a new opinion aggregation function. Extensive experiments show that these rules and the function are highly effective. A system, called Opinion Observer, has also been built.
225 citations
Proceedings Article•
01 Jan 2007TL;DR: An overview of various techniques used to tackle the problems in the domain of sentiment analysis are given, and some of their own results are added.
Abstract: The growing stream of content placed on the Web provides a huge collection of textual resources. People share their experiences on-line, ventilate their opinions (and frustrations), or simply talk just about anything. The large amount of available data creates opportunities for automatic mining and analysis. The information we are interested in this paper, is how people feel about certain topics. We consider it as a classification task: their feelings can be positive, negative or neutral. A sentiment isn’t always stated in a clear way in the text; it is often represented in subtle, complex ways. Besides direct expression of the user's feelings towards a certain topic, he or she can use a diverse range of other techniques to express his or her emotions. On top of that, authors may mix objective and subjective information about a topic, or write down thoughts about other topics than the one we are investigating. Lastly, the data gathered from the World Wide Web often contains a lot of noise. All of this makes the task of automatic recognition of the sentiment in on-line text more difficult. We will give an overview of various techniques used to tackle the problems in the domain of sentiment analysis, and add some of our own results.
201 citations
28 May 2008
TL;DR: A novel approach based on Support Vector Machines is proposed to classify a subset of documents using polarity metrics by applying it to a publicly available set of movie reviews.
Abstract: With the ever-growing popularity of online media such as blogs and social networking sites, the Internet is a valuable source of information for product and service reviews. Attempting to classify a subset of these documents using polarity metrics can be a daunting task. After a survey of previous research on sentiment polarity, we propose a novel approach based on Support Vector Machines. We compare our method to previously proposed lexical-based and machine learning (ML) approaches by applying it to a publicly available set of movie reviews. Our algorithm will be integrated within a blog visualization tool.
167 citations
05 Apr 2006
TL;DR: Simple techniques based on comparing corpus frequencies, coupled with large quantities of data, are shown to be effective for identifying the events underlying changes in global moods.
Abstract: We describe a method for discovering irregularities in temporal mood patterns appearing in a large corpus of blog posts, and labeling them with a natural language explanation. Simple techniques based on comparing corpus frequencies, coupled with large quantities of data, are shown to be effective for identifying the events underlying changes in global moods.
141 citations
04 Jun 2007
TL;DR: A survey into the scope and utility of opinion mining in legal Weblogs (a.k.a. blawgs) is performed and a language modeling approach deployed for both subjectivity analysis and polarity analysis is examined.
Abstract: We perform a survey into the scope and utility of opinion mining in legal Weblogs (a.k.a. blawgs). The number of 'blogs' in the legal domain is growing at a rapid pace and many potential applications for opinion detection and monitoring are arising as a result. We summarize current approaches to opinion mining before describing different categories of blawgs and their potential impact on the law and the legal profession. In addition to educating the community on recent developments in the legal blog space, we also conduct some introductory opinion mining trials. We first construct a Weblog test collection containing blog entries that discuss legal search tools. We subsequently examine the performance of a language modeling approach deployed for both subjectivity analysis (i.e., is the text subjective or objective?) and polarity analysis (i.e., is the text affirmative or negative towards its subject?). This work may thus help establish early baselines for these core opinion mining tasks.
108 citations