With the development of online social network platforms and social cataloging applications, large amounts of datasets are being generated daily in the form of users' reviews, evaluations, and instant messages. Readers of literary books from around the world now use social media to express their thoughts and feelings about literary works. Collecting and analyzing textual data to gain insight about the readers' interest poses a huge challenge to literary scholars and publishing industries. In this study, the authors aim to apply text analytics methods to analyze and interpret reader responses in the form of book reviews. To this end, they focus on readers' responses and reviews to Yōko Ogawa's The Housekeeper and the Professor (2003) as documented in Goodreads, a social cataloging website that allows readers across the globe to interact with each other about books. The collected data are preprocessed and explored and visualized to gain insight on public opinion about the novel. Finally, the authors analyze the collected data on Goodreads platform by using topic modelling and sentiment analysis in this chapter.
TopIntroduction
Text mining technology is now widely applicable to a wide range of industries, markets, social media, research and business needs (Rejito et al., 2021). Industries may use text mining to analyze a process and track relevant information to their daily activities. Text mining is an artificial intelligence (AI) technology that uses Natural Language Processing (NLP) to transform the free (unstructured) text in the documents and databases into normalized, structured data that can be analyzed to drive machine learning algorithms. Text mining identifies facts, relationships and assertions that would otherwise stay hidden in the mass of big data. Once extracted, the information is translated to a structured form that can be further analyzed or directly used for clustering or classification (NLP) (Malik et al., 2021; Rejito et al., 2021; Tan, 1999; Ying et al., 2021, 2022).
Text mining is also known as text data mining which is similar to text analytics. It is the process of extracting some important information from unstructured text documents such as books, websites, emails, reviews and articles (Rejito et al., 2021). Text mining is different from other available methods utilized for web searches. It is a more complex task than data mining as it involves dealing with the text data that are permanently unstructured and fuzzy (Rejito et al., 2021; Tan, 1999). Typically, text mining task includes information retrieval, text analysis, information extraction, text clustering, text categorization, visualization and data mining. The purpose of text mining is to discover a piece of unknown information or something that no one knows and could not have yet written down. A most common example of data mining in marketing and sales is banks that use data mining to analyze the use of credit card usage for the customers to purchase something. When certain outliers perform certain transactions, it also analyses fraud detection.
This study examines the Goodreads platform, an online book cataloguing website to extract some important information such as reader reviews and ratings from certain topics. Goodreads is a popular social book reading platform that allows people to rate books, post reviews, and their opinion with others. Numeous research has been conducted to examine and analyze the textual data on Goodreads. This paper seeks to analyze book readers’ reviews on Goodreads platform. In this paper, we aim to study the different methods of Natural Learning Processing (NLP) algorithms such as Topic Modelling and Sentiment Analysis. NLP is making a lot of progress in doing small subtasks in text analysis. In the present study, we use NLP to extract data of the readers’ reviews of the novel Yōko Ogawa’s The Housekeeper and the Professor (2003) from Goodreads platform (Lin & Horng, 2010). Then, the collected data is explored, normalized, and visualized to gain insight on public opinion about Yōko Ogawa’s The Housekeeper and the Professor (2003). The data is analyized by using Topic Modelling and Sentiment Analysis.