Sentiment analysis can extract information from many text sources such as reviews, news, and blogs; then it classifies them based on their polarity. Moreover, big data is produced via mobile networks and social media. Applications of sentiment analysis on big data are used as a way of classifying the opinions into diverse sentiment. Accordingly, performing sentiment analysis on big data can be helpful for a business to take useful commercial insights from text‐oriented content. However, there are very few comprehensive investigations and profound argument in this context. The goal of this paper is to provide a comprehensive and systematic investigation of the state‐of‐the‐art techniques and highlight the directions for future research. In this paper, we used systematic literature review method and in the first step, we obtained 15 351 articles; then, based on different filters, 48 related articles were attained. We have selected 23 articles based on the year of publication, the relevance of the journal, the completeness of the text, the nonrepeatability of the title, and the page number. Also, we have categorized big data and sentiment analysis into two classifications: centralized and distributed platforms. Furthermore, the disadvantages and advantages of the investigated techniques are studied and their key issues are emphasized. Consequently, this study shows that a better analysis of textual big data in terms of sentiment increases efficiency, flexibility, and intelligence. By providing comparative information and analyzing the current developments in this area, this paper will directly support academics and practicing professionals for better handling of big data in the field of sentiment analysis. This study sheds some new light on using sentiment analysis and big data for public opinion estimation and prediction.

中文翻译：

---

大数据与情感分析：全面系统的文献综述

情感分析可以从评论、新闻、博客等多种文本源中提取信息；然后它根据它们的极性对它们进行分类。此外，大数据是通过移动网络和社交媒体产生的。情感分析在大数据上的应用被用作将意见分类为不同情感的一种方式。因此，对大数据进行情感分析有助于企业从面向文本的内容中获取有用的商业见解。然而，这方面的全面调查和深刻论证却很少。本文的目标是对最先进的技术进行全面和系统的研究，并突出未来研究的方向。在本文中，我们使用系统的文献综述方法，在第一步中，我们获得了 15 351 篇文章；然后，基于不同的过滤器，获得了48篇相关文章。我们根据出版年份、期刊的相关性、文本的完整性、标题的不可重复性和页码选择了 23 篇文章。此外，我们将大数据和情感分析分为两类：集中式平台和分布式平台。此外，研究了所研究技术的缺点和优点，并强调了它们的关键问题。因此，这项研究表明，在情感方面对文本大数据进行更好的分析可以提高效率、灵活性和智能。通过提供比较信息和分析该领域的当前发展，本文将直接支持学术界和从业人员在情感分析领域更好地处理大数据。这项研究为使用情感分析和大数据进行舆情估计和预测提供了一些新的思路。