Big Data Contextual Analytics Study on Arabic Tweets Summarization

Abstract

Twitter represents a source of information as well as a free space for people to express their opinions on diverse topics. The use of twitter is rapidly increasing and generates a massive amount of data from several types and forms, in which searching for relevant tweets in a specific topic is hard manually due to irrelevant tweets. There has been much research on English tweets for understanding context; however, in spite of the fact that the Twitter active Arabic users are over hundreds of millions, there are very limited studies that have investigated Arabic tweets to produce an automatic summarization. This article proposes a multi-conversational Arabic tweets summarization approach, with a new concept of tweet classification based on influence factor. Such an approach is able to analyze Arabic tweets and provide a readable, informative, precise, concise, and diversified summary. The evaluation metrics of precision, recall, and f-measure have shown good results of the system compared to related Arabic summarization studies.