An improved evidence-based aggregation method for sentiment analysis

Abstract

Sentiment analysis is one of the natural language processing tasks used to find reviews expressed in online texts and classify them into different classes. One of the most important factors affecting the efficiency of sentiment analysis methods is the aggregation algorithm used for scores combination. Recently, Dempster–Shafer algorithm has been used for scores aggregation. This algorithm has a higher precision than common methods such as average, weighed average, product and voting, but the problem with this algorithm is the aggregation of a dominant high or low score that is always selected by the algorithm as the overall score. In the current research, a new method is proposed for scores aggregation that employs both the most and the second probable classes to predict the final score. The proposed approach considers every review as a set of sentences each of which has its own sentiment orientation and score and computes the probability of belonging of every sentence to different classes in a five-star scale using a pure lexicon-based system. These probabilities are then used for document-level sentiment detection. To this aim, two-point structure is used to improve the Dempster–Shafer aggregation algorithm. The proposed method is applied to review datasets of TripAdvisor and CitySearch which have been used in previous studies. The obtained results show that in comparison with the original Dempster–Shafer aggregation method, the precision of the proposed method for both datasets is 23% and 27% higher, respectively.

Sociocultural factors during COVID-19 pandemic: Information consumption on Twitter

“…H2 is accepted. It is clear that the use of lexicon-based methods in the messages posted on Twitter allows us to establish the level of sentiment found in each of them, which is consistent with the studies of several researchers ( Wang et al, 2019 , Jamadi Khiabani et al, 2020 , Adwan et al, 2020 , Taboada et al, 2009 ). Likewise, as suggested by Mostafa and Nebot (2017) , it was found that the use of topic recognition techniques (classification of tweets by sociocultural factors) enables us to obtain a comprehensive understanding of the underlying causes of positive or negative sentiments.…”

“…Actually, the study of the level of sentiment carried out in tweets allows us to compare the criteria of different researchers. Thus, it is demonstrated that thanks to the lexicon used in the tweets, we can understand specific ideas, judgments, concepts, and topics regarding the events of the COVID-19 pandemic ( Wang et al, 2019 , Jamadi Khiabani et al, 2020 , Taboada et al, 2009 ). In addition, when comparing the words used in the COVID-19 tweets and the list of words in the “Lexicom” dictionary that is embedded in the library (Sentiment Analysis) of the R program, we can confirm that the polarity score makes it possible to determine the positive, negative or neutral influence on the sentence ( Adwan et al, 2020 , Miao et al, 2010 ).…”

“…(2019) stated that lexicon-based methods are conducive and have received much acceptance when analyzing the ideas, judgments or concepts expressed by people in different social spaces. Jamadi et al (2020) affirmed that a lexicon-based approach is quick and independent to analyze the different topics that are in the conversations. Adwan et al.…”

Customer Sentiment Recognition in Conversation Based on Contextual Semantic and Affective Interaction Information

“…Four evaluation criteria, namely precision (Pr), recall (Re), F1measure (F1), and accuracy (Acc), are used to assess the performance of the models. These criteria are extensively used in text classification and sentiment analysis tasks [23]. These criteria are calculated as follows:…”

Dynamic PScore: A Dynamic Method to Prioritize User Reviews

“…The apps available in the store are downloaded by many users and Google Play Store users could comment on the desired application. Studies have demonstrated the reviews made by users on apps contains important information, including bug reports, feature requests and user experience of working with the app [3][4][5].…”