Scribd
Upload
55 views16 pages

Shiva

This document discusses a system for detecting fake product reviews using opinion mining and machine learning techniques. It provides an overview of the challenges of fake reviews online and the goal of creating a trustworthy environment for e-commerce. The proposed system involves preprocessing data, feature extraction, and training classification models like Naive Bayes, Random Forest, and Decision Tree to identify fake reviews. While no single algorithm is perfect, hybrid and evolving models could improve accuracy and speed. Continued research is needed to stay ahead of increasingly sophisticated fake review generation.

Uploaded by

DreamBlackAble AE
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
55 views16 pages

Shiva

This document discusses a system for detecting fake product reviews using opinion mining and machine learning techniques. It provides an overview of the challenges of fake reviews online and the goal of creating a trustworthy environment for e-commerce. The proposed system involves preprocessing data, feature extraction, and training classification models like Naive Bayes, Random Forest, and Decision Tree to identify fake reviews. While no single algorithm is perfect, hybrid and evolving models could improve accuracy and speed. Continued research is needed to stay ahead of increasingly sophisticated fake review generation.

Uploaded by

DreamBlackAble AE
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

OPINION MINING BASED FAKE PRODUCT

REVIEW MONITORING AND REMOVAL


SYSTEM
Shivathmaj Shenoy M
4CB18CS085
Agenda
Overview
Working of Fake Review detection
Fake Review Survey Using Machine learning Approach
Fake Review detection Using Naïve Bias
Fake Review detection Using Random Forest Classifier
Fake Review detection using Decision tree
Scope of Growth and Disadvantages
Conclusion
Reference
Overview
The reviews on a product may be positive or negative, the negative reviews will attract the
customers more than a positive review. These fake reviews can affect any business which
leads financial profit or loses.
Due to the financial reasons many fake reviews will be appeared in these websites. The
company owners will intentionally motivate some of the people to write the fake reviews to
improve their business towards another product.
The focus of this research is to create an environment of online E-commerce industry where
consumers build trust in a platform where the products they purchase are genuine and
feedbacks posted on these websites/applications are true
So our Final aim in this is to implement best approach available for detection of fake
reviews using opinion mining (sentiment analysis) techniques. To let users, know if each
individual review is trustworthy or not for efficient use of money from users side.
Working of Fake Review detection
1. Preprocessing
2. Dataset
3. Feature Extraction
4. Model Training
Fake Review Survey Using Machine learning Approach

Figure 3.11: Machine Learning based Fake Review Detection


Fake Review Survey Using Machine learning Approach
World Wide Web has drastically changed the way of sharing the opinions.
Online reviews are comments, tweets, and posts, opinions on different online
platforms like review sites, news sites, e-commerce sites or any other social
networking sites
1. Untruthful reviews
2. Reviews on brands
3. Review Centric Approach
4. Centric Approach
5. Product Centric Approach
Fake Review Survey Using Machine learning Approach
Data Collection
Data pre-processing
Feature Extraction and selection
Classifier model construction and testing
Working of Fake Review detection

Figure 3.1: Model Diagram for fake review detection


Fake Review Detection Using Naïve Bias

Figure 3.3: Naïve Bias Equation


Naïve Bias Vs. Random Forest Classifier

Figure 3.4: Model Diagram for fake review detection


Fake Review detection using Decision tree

Figure 3.6: Machine Learning based Fake Review


Application:
Scope of Growth and Disadvantages
 No Fixed Algorithm we need to choose a Algorithm based on Criteria and this is a manual task

 In future work, hybrid models and new models can be tried for the fake review detection model.

 Process is slow we can increase its speed by using Graphical Processing Unit

 Need for constant evolution of methods because humans always find smarter ways to cheat the system
Conclusion

Identifying fake reviews from a large dataset is challenging enough to become an


important research problem. Business organizations, specialists and academics are
battling to find the best system for opinion spam analysis.
The most important part of an algorithm is its efficiency. Efficiency is not just about
execution time. The efficiency of an algorithm is about the time taken for training the
model and the time taken for the prediction
Reference
[1]. Dhairya Patel, Aishwerya Kapoor and Sameet Sonawane “Fake review detection using opinion mining” International Research journal of Engineering and

technology (IRJET) , volume 5, issue 12,Dec 2018.

[2]. Jitendra kumar Rout, Amiya Kumar Dash and Niranjan Kumar Ray “Framework for Fake Review Detection: Issues and Challenges” IEEE Xplore ISBN:978-1-

7281-0259-7/18 2018.

[3]. Nidhi A. Patel and Prof. Rakesh Patel “A Survey on Fake Review Detection using Machine Learning Techniques” IEEE Xplore ISBN: 978-1-5386-6947-1/18

2018.

[4]. Sanjay K.S and Dr.Ajit Danti “Detection of fake opinions on online products using Decision Tree and Information Gain” IEEE Xplore ISBN: 978-1-5386-7808-4

2019.

[5]. Syed Mohammed Anas and Santoshi Kumari “Opinion Mining based Fake Product review Monitoring and Removal System” IEEE Xplore ISBN: 978-1-7281-

8501-9 2021.
THANK YOU

You might also like