Aspect-based Sentiment Analysis for Bengali Text using Bidirectional Encoder Representations from Transformers (BERT)

Abstract

Public opinion is important for decision-making on numerous occasions for national growth in democratic countries like Bangladesh, the USA, and India. Sentiment analysis is a technique used to determine the polarity of opinions expressed in a text. The more complex stage of sentiment analysis is known as Aspect-Based Sentiment Analysis (ABSA), where it is possible to ascertain both the actual topics being discussed by the speakers as well as the polarity of each opinion. Nowadays, people leave comments on a variety of websites, including social networking sites, online news sources, and even YouTube video comment sections, on a wide range of topics. ABSA can play a significant role in utilizing these comments for a variety of objectives, including academic, commercial, and socioeconomic development. In English and many other popular European languages, there are many datasets for ABSA, but the Bengali language has very few of them. As a result, ABSA research on Bengali is relatively rare. In this paper, we present a Bengali dataset that has been manually annotated with five aspects and their corresponding sentiment. A baseline evaluation was also carried out using the Bidirectional Encoder Representations from Transformers (BERT) model, with 97% aspect detection accuracy and 77% sentiment classification accuracy. For aspect detection, the F1-score was 0.97 and for sentiment classification, it was 0.77.

Public Perception of Amil Zakat Institutions in Indonesia: Insight Discovery From Machine Learning

“…Labeling will be given to a group of data with negative, neutral, or positive labels based on sentiment scores (Asmat et al, 2022). Some previous studies, such as Hamid et al, (2022), Hudaefi, (2021) used sentiment analysis to measure people's perceptions of Islamic social donations. Based on this background, the questions to be answered by this research are: RQ3: What are the sentiments of netizens on Twitter regarding the role and performance of the Amil Zakat institution?…”

“…Some other studies, such as Assalafiyah & Rusydiana, (2023) aim to measure sentiment about zakat but use sources derived from secondary data such as reputable scientific journals. At the same time, Hamid et al, (2022) tried to measure public satisfaction regarding zakat with a sentiment analysis approach. On the other hand, this research has a fundamental difference in terms of the object used, where this research will focus more on measuring sentiment and perception in zakat fund management organizations or amil zakat institutions.…”

A Novel Approach: Tokenization Framework based on Sentence Structure in Indonesian Language

“…To be processed, text data needs to be prepared at the textpre-processing stage. This stage is the first important step of any data mining process to achieve better accuracy [3]. This process will change the data from its original form into a form that is easier to observe and explore [4].…”

Research on emotion analysis of e-commerce product reviews based on deep learning

“…Bert is one of the pre training language models, and its model structure is shown in Figure 1. It can be seen from Figure 1 [12][13] .GRU is the same as LSTM, both of which are currently highly used gated recurrent neural networks. Structurally, LSTM uses three gates for information flow, and only two gates in GRU are used for control, namely reset and update.…”