Periodicals of Engineering and Natural Sciences ISSN 2303-4521 Vol. 7, No. 1, June 2019, pp.420-432 Available online at: http://pen.ius.edu.ba 420 Review Helpfulness Prediction: Survey Yasamyian Almutairi 1 , Manal Abdullah 2 , Dimah Alahmadi 3 1 Master student in Information system at king Abduziz university 2 Associated Professor in king Abuduziz university 3 Assistant Professor in king Abuduziz university Article Info ABSTRACT Received Dec 28 th , 2018 Online reviews have become the major driving factor influencing purchasing behavior and patterns of social customers. However, it is difficult for customer to cover good reviews about any product or service according to massive amount of reviews latest years. Many previous researches provide innovative models about predicting review helpfulness in E-commerce websites. Some of these studies exploring the direct effect of review attributes on review helpfulness while others focused on reviewer’s attributes only. The main objective of this research is to review the most important attributes that have an affect on review helpfulness from many perspectives such as datasets, techniques, frameworks and evaluation methods of the experiments. The paper ends up with important findings about most attributes effect the review helpfulness such as Review Valence. Keyword: Review helpfulness, Recommender system Corresponding Author: Second Author, Departement of Information System, King Abduaziz University, Jeddah, Saudi Arabia, Email: maaabdullah@kau.edu.sa. 1. Introduction Online reviews are become important factor in assisting customers’ buying decisions. Reviews offer valuable information can influence customers’ opinion. Moore in [1] states that 92% of customers nowadays read online reviews. This makes online review helpfulness more key factor in E-commerce platform. Moreover, online reviews differ in their support to customers due to different subjective. For example, some of customers discover the reviews that support their decision-making such as product evaluation reviews. In other words, they looks for its utility that called “review helpfulness” [2]. Review helpfulness indicates whether the review gives useful product assessment and buying decision to other customers. Hence, it is important to explore attributes and which make the review more helpful. These attributes belong to two major categories: the first is about the review itself such as review length, rating valence, and review extremity [ 3 ] . The second is related to reviewers such as reviewer Social Profile Information (SPI), reviewer ranking, and reviewer engagement [4]. Due to the importance of the helpfulness of reviews to the social customers and the numerous profits to the E- commerce website, this research paper discusses the main perspectives of reviews helpfulness prediction. In next section, we will identify the concept of review helpfulness in E-commerce websites then illustrate the important attributes that affect the review helpfulness. The next sections are revised of dataset, techniques, frameworks and evaluation methods of the review helpfulness prediction.