به سوی چارچوب داده بزرگ برای تجزیه و تحلیل محتوای رسانه های اجتماعی
Abstract
1- Introduction
2- Background
3- Data integration and big data analytics framework
4- Results
5- Discussion
6- Conclusions
References
Abstract
Modern companies generate value by digitalizing their services and products. Knowing what customers are saying about the firm through reviews in social media content constitutes a key factor to succeed in the big data era. However, social media data analysis is a complex discipline due to the subjectivity in text review and the additional features in raw data. Some frameworks proposed in the existing literature involve many steps that thereby increase their complexity. A two-stage framework to tackle this problem is proposed: the first stage is focused on data preparation and finding an optimal machine learning model for this data; the second stage relies on established layers of big data architectures focused on getting an outcome of data by taking most of the machine learning model of stage one. Thus, a first stage is proposed to analyze big and small datasets in a non-big data environment, whereas the second stage analyzes big datasets by applying the first stage machine learning model of. Then, a study case is presented for the first stage of the framework to analyze reviews of hotel-related businesses. Several machine learning algorithms were trained for two, three and five classes, with the best results being found for binary classification.
Introduction
Social media companies became popular with the advent of the Internet in the late 1990s. In those early days, users expressed their feelings about the products they bought or the services they used commonly through blogs, web chats in dedicated forums or via email to the provider. As e-commerce continued evolving, enterprises such as Amazon and the Internet Movie Database (IMDb) included for every item (e.g. CDs, books, DVDs, movies, TV series, etc.) a means for registered users to be able to interact among themselves and to share opinions about their buying experiences. Since then, these services have evolved in many ways to offer users more sophisticated methods to enrich the review experience. Some of the add-ons that now come along with the review text are: number of stars on a given scale, number of votes that found the review useful, photo of the reviewer, popularity of the reviewer, number of reviews given by the reviewer, images to illustrate or support the argument, kind of services provided (indicated by the customers), overall rating of the service/product provider, etc. Many of the features mentioned above have been integrated into services by digital companies such as TripAdvisor, Airbnb, Amazon, Yelp, Cabify, Blablacar, Foursquare and Booking.com. These features generate giant volumes of information that are commonly referred to as Big Data (BD): Petabytes and even exabytes of data that are being generated by these type of enterprises (Gandomi & Haider, 2015). Companies of a minor scale not solely dedicated to digital services are also generating big volumes of data that reach terabytes of data on a regular basis. For further information, Yaqoob et al. (2016) present a robust study of the evolution of BD from its conception to its future challenges, aimed at a more comprehensive understanding of the BD scenario.