Review-Based Recommender Systems: A Proposed Rating Prediction Scheme Using Word Embedding Representation of Reviews

2020 ◽  
Author(s):  
S Hasanzadeh ◽  
S M Fakhrahmad ◽  
M Taheri
Keyword(s):  
Vector Space ◽  
Recommender Systems ◽  
Recommender System ◽  
Word Embedding ◽  
Bag Of Words ◽  
Prediction System ◽  
User Reviews ◽  
Prediction Scheme ◽  
Rating Prediction ◽  
The Given

Abstract Recommender systems nowadays play an important role in providing helpful information for users, especially in ecommerce applications. Many of the proposed models use rating histories of the users in order to predict unknown ratings. Recently, users’ reviews as a valuable source of knowledge have attracted the attention of researchers in this field and a new category denoted as review-based recommender systems has emerged. In this study, we make use of the information included in user reviews as well as available rating scores to develop a review-based rating prediction system. The proposed scheme attempts to handle the uncertainty problem of the rating histories, by fuzzifying the given ratings. Another advantage of the proposed system is the use of a word embedding representation model for textual reviews, instead of using traditional models such as binary bag of words and TFIDF 1 vector space. It also makes use of the helpfulness voting scores, in order to prune data and achieve better results. The effectiveness of the rating prediction scheme as well as the final recommender system was evaluated against the Amazon dataset. Experimental results revealed that the proposed recommender system outperforms its counterparts and can be used as a suitable tool in ecommerce environments.

scholarly journals GENERATING ARTIFICIAL DATA FOR SCALABILITY TEST OF A FOLLOWEE TWITTER RECOMMENDER SYSTEM

2021 ◽  
Author(s):  
Jason Li
Keyword(s):  
Recommender Systems ◽  
Recommender System ◽  
Analytical Modeling ◽  
Real Data ◽  
Bag Of Words ◽  
Development Phase ◽  
Artificial Data ◽  
Distributed Environment ◽  
Privacy Concerns ◽  
Complete Development

The current methods for the evaluation of the scalability of recommender systems measure the scalability of the whole application after deployment on a cloud by calculating the running times of the application when increasing the number of nodes. This method requires the complete development and implementation of a whole application. To be able to test the recommender system during the development phase, the major problem to test the scalability and accuracy is collecting real data (i.e. social data), which is a time-consuming task and sometimes it is not possible due to privacy concerns. This thesis proposes measuring the scalability of Twitter recommender systems by simulating the software, which processes a large number of artificial tweets. A method is introduced and validated for producing artificial tweets to test a recommender system. This method of producing artificial tweets is based on using analytical modeling, tf-idf and bag-of-words model. A simulator is developed to test the scalability of a recommender system and underlying distributed environment.

A Hybrid Recommender System for Rating Prediction of Arabic Reviews

2020 ◽  
pp. 2050010
Author(s):  
Fouzi Harrag ◽  
Abdulmalik Salman Al-Salman ◽  
Alaa Alquahtani
Keyword(s):  
Sentiment Analysis ◽  
Recommender Systems ◽  
Recommender System ◽  
Opinion Mining ◽  
Good Accuracy ◽  
Hybrid Approach ◽  
Data Sparsity ◽  
Hybrid Recommender System ◽  
Rating Prediction ◽  
Hybrid Recommender

Recommender systems nowadays are playing an important role in the delivery of services and information to users. Sentiment analysis (also known as opinion mining) is the process of determining the attitude of textual opinions, whether they are positive, negative or neutral. Data sparsity is representing a big issue for recommender systems because of the insufficiency of user rating or absence of data about users or items. This research proposed a hybrid approach combining sentiment analysis and recommender systems to tackle the problem of data sparsity problems by predicting the rating of products from users’ reviews using text mining and NLP techniques. This research focuses especially on Arabic reviews, where the model is evaluated using Opinion Corpus for Arabic (OCA) dataset. Our system was efficient, and it showed a good accuracy of nearly 85% in predicting the rating from reviews.

scholarly journals Designing and Implementation of Rating Prediction

2019 ◽  
Vol 8 (11) ◽  
pp. 2847-2852
Keyword(s):  
Recommender Systems ◽  
Recommender System ◽  
Rating System ◽  
The Real ◽  
Fine Grained ◽  
Elo Rating System ◽  
Rating Prediction ◽  
Farthest Point

Recommender system is one of indivisible parts in web business areas. Recommender system construes the system which underwrites things for the client who wish to purchase things .One of the real inconveniences that, figuratively speaking, stays in recommender system is the virus begin problem(inactive things) which can be seen as an obstruction that spurns the cool begin things from the present things. In this paper, we want to move beyond this farthest point for cold-begin clients/things by the help of existing things. It is developed by utilizing Elo Rating system. The Elo system is widely gotten in chess competitions; we propose a novel rating association technique to get settled with the profiles of cold-begin things. The purpose of assembly of our Strategy is to give a fine-grained View on the shrouded profiles of cold-begin clients/things by inspecting the separations between nippy begin things and existing Products. To uncover the limit of methodology, we instantiate our technique on two typical strategies in recommender systems, i.e., the structure factorization based, and neighborhood based pleasing sifting. Starter assessments on five genuine instructive documents embrace the amazing quality of our methodology over the present procedures in virus begin situation

scholarly journals Measuring user’s influence in the Yelp recommender system

2017 ◽  
Vol 1 (2) ◽  
pp. 91-104 ◽  
Author(s):  
Andres Bejarano ◽  
Agrima Jindal ◽  
Bharat Bhargava
Keyword(s):  
Recommender Systems ◽  
Recommender System ◽  
Design Methodology ◽  
User Reviews ◽  
Content Type ◽  
Attribute Weights ◽  
User Influence ◽  
Proposed Model ◽  
The Social

Purpose Recommender systems collect information about users and businesses and how they are related. Such relation is given in terms of reviews and votes on reviews. User reviews gather opinions, rating scores and review influence. The latter component is crucial for determining which users are more relevant in a recommender system, that is, the users whose reviews are more popular than the average user’s reviews. Design/methodology/approach A model of measure of user influence is proposed based on review and social attributes of the user. User influence is also used for determining how influenced has been a business being based on popular reviews. Findings Results indicate there is a connection between social attributes and user influence. Such results are relevant for marketing, credibility estimation and Sybil detections, among others. Originality/value The proposed model allows search parameterization based on the social attribute weights of users, reviews and businesses. Such weights defines the relevance on each attribute, which can be adjusted according to the search needs. Popularity results are then a function of weight preferences on user, reviews and businesses data join.

scholarly journals GENERATING ARTIFICIAL DATA FOR SCALABILITY TEST OF A FOLLOWEE TWITTER RECOMMENDER SYSTEM

2021 ◽  
Author(s):  
Jason Li
Keyword(s):  
Recommender Systems ◽  
Recommender System ◽  
Analytical Modeling ◽  
Real Data ◽  
Bag Of Words ◽  
Development Phase ◽  
Artificial Data ◽  
Distributed Environment ◽  
Privacy Concerns ◽  
Complete Development

The current methods for the evaluation of the scalability of recommender systems measure the scalability of the whole application after deployment on a cloud by calculating the running times of the application when increasing the number of nodes. This method requires the complete development and implementation of a whole application. To be able to test the recommender system during the development phase, the major problem to test the scalability and accuracy is collecting real data (i.e. social data), which is a time-consuming task and sometimes it is not possible due to privacy concerns. This thesis proposes measuring the scalability of Twitter recommender systems by simulating the software, which processes a large number of artificial tweets. A method is introduced and validated for producing artificial tweets to test a recommender system. This method of producing artificial tweets is based on using analytical modeling, tf-idf and bag-of-words model. A simulator is developed to test the scalability of a recommender system and underlying distributed environment.

Collaborating The Textual Reviews Of The Merchandise and Foretelling The Rating Supported Social Sentiment

2021 ◽  
pp. 63-72
Author(s):  
Vijay K ◽  
Keyword(s):  
Recommender Systems ◽  
Recommender System ◽  
Geographic Area ◽  
Comprehensive Analysis ◽  
User Preferences ◽  
Significant Information ◽  
The Social ◽  
Prediction Technique ◽  
Considerable Length ◽  
Rating Prediction

Lately, we have seen a twist of audit sites. It presents a decent opportunity to share our experience for a considerable length of time we have bought. Be that as it may, we tend to confront the information over-burdening issue. A method for mining significant information from surveys to know a client's inclinations and produce precise proposal is fundamental. Since quite a while ago settled recommender Systems (RS) considers a few variables, similar to client's buy records, item class, and geographic area. During this work, we have proposed sentiment-based rating prediction technique (RPS) to help up the expectation precision in recommender Systems. First and foremost, we examine the social user sentimental measuring approach and calculate every user’s sentiment on things/items. Furthermore, we don't exclusively consider a client's own wistful properties anyway moreover take interpersonal social sentimental influence into study. Then, at that point, we propose to consider item name, which might be deduced by the sentimental distributions of a user set that reflect clients' comprehensive analysis. Finally, we tend to intertwine 3 factors-user sentiment similarity, interpersonal social sentimental distributions of a client opinion likeness, interpersonal social sentimental influence, associate the thing's reputation relationship into our recommender system to make a talented rating prediction. Then, at that point, we arranged a presentation analysis of the 3 sentimental factors on a genuine world dataset gathered from Yelp. Our exploratory outcomes show, the sentiment will well describe user preferences, which facilitate to hike the proposal execution.

scholarly journals A Systematic Review of Recommender Systems and Their Applications in Cybersecurity

Sensors ◽  
2021 ◽  
Vol 21 (15) ◽  
pp. 5248
Author(s):  
Aleksandra Pawlicka ◽  
Marek Pawlicki ◽  
Rafał Kozik ◽  
Ryszard S. Choraś
Keyword(s):  
Systematic Review ◽  
Recommender Systems ◽  
Recommender System ◽  
State Of The Art ◽  
The State ◽  
Advantages And Disadvantages ◽  
Comprehensive Survey ◽  
Security Concerns ◽  
Valuable Role

This paper discusses the valuable role recommender systems may play in cybersecurity. First, a comprehensive presentation of recommender system types is presented, as well as their advantages and disadvantages, possible applications and security concerns. Then, the paper collects and presents the state of the art concerning the use of recommender systems in cybersecurity; both the existing solutions and future ideas are presented. The contribution of this paper is two-fold: to date, to the best of our knowledge, there has been no work collecting the applications of recommenders for cybersecurity. Moreover, this paper attempts to complete a comprehensive survey of recommender types, after noticing that other works usually mention two–three types at once and neglect the others.

Modelling trust networks using resistive circuits for trust-aware recommender systems

2016 ◽  
Vol 43 (1) ◽  
pp. 135-144 ◽  
Author(s):  
Mehdi Hosseinzadeh Aghdam ◽  
Morteza Analoui ◽  
Peyman Kabiri
Keyword(s):  
Collaborative Filtering ◽  
Recommender Systems ◽  
Recommender System ◽  
Cold Start ◽  
User Preferences ◽  
Item Selection ◽  
Trust Networks ◽  
Trust Inference ◽  
Trust Information

Recommender systems have been widely used for predicting unknown ratings. Collaborative filtering as a recommendation technique uses known ratings for predicting user preferences in the item selection. However, current collaborative filtering methods cannot distinguish malicious users from unknown users. Also, they have serious drawbacks in generating ratings for cold-start users. Trust networks among recommender systems have been proved beneficial to improve the quality and number of predictions. This paper proposes an improved trust-aware recommender system that uses resistive circuits for trust inference. This method uses trust information to produce personalized recommendations. The result of evaluating the proposed method on Epinions dataset shows that this method can significantly improve the accuracy of recommender systems while not reducing the coverage of recommender systems.

scholarly journals Explaining Review-Based Recommendations: Effects of Profile Transparency, Presentation Style and User Characteristics

i-com ◽  
2020 ◽  
Vol 19 (3) ◽  
pp. 181-200
Author(s):  
Diana C. Hernandez-Bocanegra ◽  
Jürgen Ziegler
Keyword(s):  
Decision Making ◽  
Individual Differences ◽  
User Interface ◽  
Recommender Systems ◽  
Social Awareness ◽  
User Characteristics ◽  
Presentation Style ◽  
User Reviews ◽  
Decision Making Styles

Abstract Providing explanations based on user reviews in recommender systems (RS) may increase users’ perception of transparency or effectiveness. However, little is known about how these explanations should be presented to users, or which types of user interface components should be included in explanations, in order to increase both their comprehensibility and acceptance. To investigate such matters, we conducted two experiments and evaluated the differences in users’ perception when providing information about their own profiles, in addition to a summarized view on the opinions of other customers about the recommended hotel. Additionally, we also aimed to test the effect of different display styles (bar chart and table) on the perception of review-based explanations for recommended hotels, as well as how useful users find different explanatory interface components. Our results suggest that the perception of an RS and its explanations given profile transparency and different presentation styles, may vary depending on individual differences on user characteristics, such as decision-making styles, social awareness, or visualization familiarity.

Sign in / Sign up

Export Citation Format

Share Document