Review Spam Detection using Machine Learning

ABSTRACT:

Prior to buying a product, people usually inform themselves by reading online reviews. To make more profit sellers often try to fake user experience. As customers are being deceived this way, recognizing and removing fake reviews is of great importance. This paper analyzes spam detection methods, based on machine learning, and presents their overview and results.

PROJECT OUTPUT VIDEO:

EXISTING SYSTEM:

• Researchers observed that many spammers copy existing reviews entirely or change only a few words. Therefore, many researchers in this area have focused on methods for duplicates detection. These methods are used for finding textual or conceptual similarity between reviews.

• In existing authors used conceptual similarity. They used data collected from a digital camera page and extracted its main features (photo quality, design, zoom, size etc.). Then for each review they extracted which features were mentioned and in which context. Using those features they calculated similarities between reviews and compared them. Using labels obtained by two human observers as ground truth, their method achieved only 43.6% accuracy.

DISADVANTAGES OF EXISTING SYSTEM:

• Untruthful opinions represent purposefully fake reviews. Reviews on brands only aren’t focused on products, but rather on brands or manufacturers.

• Non-reviews include advertisements or other irrelevant reviews containing no opinions. Although types two and three fail to address specific products, they aren’t fraudulent.

• These types of spam are also easy to spot manually and traditional classification approaches have no problem in detecting them.

• Untruthful reviews are shown to be much harder task for a machine as well as for a human observer.

PROPOSED SYSTEM:

• In the proposed system, we mainly focus on characterize users and their rating behaviors and the helpfulness scores received from others and the correlation of their reviews with product popularity.

• In this system we implement the system of spam detection methods is presented which uses machine learning algorithm. It was shown that using different datasets yields extremely different results.

ADVANTAGES OF PROPOSED SYSTEM:

• A higher average rating score of spam detection

• A higher helpfulness score of spam detection is likely to increase or decrease product popularity

MODULES:

• Data Collection
• Admin
• Spam Review Detection

MODULE DESCRIPTIONS:

Data Collection:

Dataset used in this process, contain reviews from tripadvisor, Amazon Mechanical Turk, and Web Sources. Each Dataset consists of positive and negative polarity reviews. Each positive and negative polarity directory consists of fake and ham (true reviews).

Admin:

In this Module, Admin can manage their website such as uploading dataset which contain reviews from their Website. The important feature of the Admin is to find the spam Reviews Posted on the Website about their service.

Spam Review Detection:

By the training of the system, the system will categorize the spam reviews by their characteristics such as:

1. Comments without alphanumerical characters

2. Random combination of characters

3. Random combination of words

4. Common phrases

5. Links

6. Text with links

SYSTEM REQUIREMENTS:

HARDWARE REQUIREMENTS:

• System           : Pentium Dual Core.

• Hard Disk      : 120 GB.

• Monitor          : 15’’ LED

• Input Devices : Keyboard, Mouse

• Ram                : 1 GB

SOFTWARE REQUIREMENTS:

• Operating system : Windows 7.

• Coding Language : Python

• Database               : MYSQL

REFERENCE:

Draško Radovanović, Božo Krstajić, Member, IEEE, “Review Spam Detection using Machine Learning” 2018 23rd International Scientific-Professional Conference on Information Technology (IT), 2018.