Feature-level Sentiment Research Articles

ONTOLOGY-BASED APPROACH FOR FEATURE LEVEL SENTIMENT ANALYSIS

Through state-of-the-art digitalization, we can see a massive growth in user generated content on the web that provides feedback from people on a variety of topics. However, manually managing large-scale user feedback would be a difficult task and a waste of time. Therefore, the concept of sentiment analysis is emerged. Sentiment analysis is a computerized study of individuals' feelings and opinions about an entity or product. It can be executed at three different levels: document level, sentence or phrase level, and feature level. This paper proposes a novel ontology-based approach for feature level sentiment analysis. The proposed approach extracts the product features using semantic similarity and Wordnet ontology and uses the SentiWordent dictionary to classify the users’ comments as positive and negative. Furthermore, it manages negative words to obtain more precise classification results. The proposed approach is assessed by using two benchmark amazon products’ datasets in terms of accuracy; recall, precision, and f-measure. The performance reaches to 92.4% accuracy, 97.2% precision, 92.8 % recall, and 94.4% f-measure.

Asymmetric effect of feature level sentiment on product rating: an application of bigram natural language processing (NLP) analysis

PurposeThe evaluation of perceived attribute performance reflected in online consumer reviews (OCRs) is critical in gaining timely marketing insights. This study proposed a text mining approach to measure consumer sentiments at the feature level and their asymmetric impacts on overall product ratings.Design/methodology/approachThis study employed 49,130 OCRs generated for 14 wireless earbud products on Amazon.com. Word combinations of the major quality dimensions and related sentiment words were identified using bigram natural language processing (NLP) analysis. This study combined sentiment dictionaries and feature-related bigrams and measured feature level sentiment scores in a review. Furthermore, the authors examined the effect of feature level sentiment on product ratings.FindingsThe results indicate that customer sentiment for product features measured from text reviews significantly and asymmetrically affects the overall rating. Building upon the three-factor theory of customer satisfaction, the key quality dimensions of wireless earbuds are categorized into basic, excitement and performance factors.Originality/valueThis study provides a novel approach to assess customer feature level evaluation of a product and its impact on customer satisfaction based on big data analytics. By applying the suggested methodology, marketing managers can gain in-depth insights into consumer needs and reflect this knowledge in their future product or service improvement.

Opinion Extraction on Online Malay Text

Growing of social media usage present a new set of opportunities and challenges in the way of information is retrieved and searched. Opinions on social media has become an important factor in influencing people choices on purchasing a product and service. Hence, sentiment analysis has become the most crucial tool in tracking people feedbacks on products and services. For Malay language there is limited sources available for this language. Thus, in this paper we present the method of extracting opinion on online Malay text. The traditional method using POS extraction is not adequate. Thus, rule based method is integrated with POS extraction method to improve opinion words extraction. Most of the existing tools are able to retrieve opinion at sentence and document level. More detail analysis is acquired to have detail information and summarization of a product. This is where feature level sentiment analysis is needed. The process of identifying opinion of a particular feature in a sentence, can be quite tedious and troublesome. This is because opinion of the feature can be hidden and scattered in the sentence. Therefore, opinion mapping is employed for opinion extraction at feature level in this paper. A set of tweets from telecommunication domain is used to evaluate the proposed framework. From the experiment, the accuracy of the extraction performed is 88%. The detail description of the feature level opinion extraction steps is discussed in this paper.

Consumer insight mining: Aspect based Twitter opinion mining of mobile phone reviews

Micro-blogging sites such as Twitter are often considered as rich source of opinions of the masses towards products. The character length limit in tweets encourages people to use emojis, emoticons and out of vocabulary words. Due to the huge volume of tweets being generated, it is difficult to manually label tweets and create a supervised learning model for sentiment analysis. Looking into these challenges, the research paper aims to create a feature level sentiment analysis model for Twitter data mining including features such as emoji detection, spelling correction and emoticon detection. The proposed model consists of automated training data labeling by using lexicon based approach. It is an ontology based system with the domain of “Smartphone”. In addition to the general lexicon used, a set of lexicons specific for each attribute of the domain “Smartphone” are used to improve classification accuracy for training data generation. This is used to classify tweets obtained about a particular mobile phone using SVM classifier. Experimental results show that the classifier based on automated training data provides good accuracy. It also demonstrates the importance of emoji detection and the attribute specific lexicons which help improve the classification accuracy.

Open source software adoption evaluation through feature level sentiment analysis using Twitter data

Adopting open source software from the Internet, developers often encounter the problem of accessing the quality of candidate software. To efficiently adopt the system they need a sort of quality guarantee regarding software resources. To assist the developer in software adoption evaluation we have proposed a software adoption assessment approach based on user comments. In our proposed approach, we first collected the textual reviews regarding the software resource, assigned the sentiment polarity (positive or negative) to each comment, extracted the adoption aspect which the comment talks about, and then based on the adoption aspects of the software generated an aggregated sentiment profile of the software. Twitter micro-blogging data about OSS products were crawled, preprocessed, tagged, and then summarized. To evaluate the proposed model, a set of experiments was designed and conducted using different classifiers, i.e. Apriori, GSP, and AdaBoost. For the feature level sentiment summarization we have used Bayesian statistics and frequency distribution techniques. The results show that the proposed approach achieved satisfying precision and recall, i.e. above 80% along with an average accuracy of 70.98%.

A comparative performance evaluation of neural network based approach for sentiment classification of online reviews

The aim of sentiment classification is to efficiently identify the emotions expressed in the form of text messages. Machine learning methods for sentiment classification have been extensively studied, due to their predominant classification performance. Recent studies suggest that ensemble based machine learning methods provide better performance in classification. Artificial neural networks (ANNs) are rarely being investigated in the literature of sentiment classification. This paper compares neural network based sentiment classification methods (back propagation neural network (BPN), probabilistic neural network (PNN) & homogeneous ensemble of PNN (HEN)) using varying levels of word granularity as features for feature level sentiment classification. They are validated using a dataset of product reviews collected from the Amazon reviews website. An empirical analysis is done to compare results of ANN based methods with two statistical individual methods. The methods are evaluated using five different quality measures and results show that the homogeneous ensemble of the neural network method provides better performance. Among the two neural network approaches used, probabilistic neural networks (PNNs) outperform in classifying the sentiment of the product reviews. The integration of neural network based sentiment classification methods with principal component analysis (PCA) as a feature reduction technique provides superior performance in terms of training time also.

Feature Ranking in Sentiment Analysis

With the rapid expansion of e-commerce over the past 15 years, more products are sold on the Web. More and more people are buying products online. In order to enhance customer shopping experience, it has become a common practice for online merchants to enable their customers to write reviews on products that they have purchased. Some popular products can get hundreds of reviews or more at some large merchant sites. Manual analysis of customer opinions is only possible to a certain extent and very time-consuming due to the multitude of contributions.From the e-commerce perspective, receiving consumer’s feedback can greatly improve its strategies in order to increase products of the sector. This research work will present feature wise sentiment analysis of customer review. The goal of feature level sentiment analysis is to produce a feature-based opinion summary of multiple reviews With summaries of opinions and features of the product, people can make effective decisions in less time. Such mining can be helpful for competitive marketing. Feature extraction can be performed using two approaches. Rule-based algorithm and HAC algorithm. Feature ranking will be done using MAX opinionscore algorithm and opinion score obtained from SentiWordNet.