The main goal of social network analysis is the study of structural properties of networks. Structural analysis of the social network investigates the properties of individual vertices and the global properties of the network as a whole. It answers two basic classes of questions about the network: what is the structural position of any given individual node and what can be said about groups forming within the network. The main measurement of a node’s social power is centrality, which allows to determine node’s relative and absolute importance in the network. There are several methods to determine node’s centrality, such as the degree centrality the betweenness centrality or the closeness centrality. The proposed system will able to mine users’ intent from comments. Irrelevant comments removal will increase opinion mining performance of system. False positive & false negative rates may reduced. Resistant to fake opinion postings.
The Essentials of Digital Experience Monitoring_ A Comprehensive Guide.pdf
Opinion Mining & Sentiment Analysis Based on Natural Language Processing
1. Ocular Systems Pune Info@ocularsystems.in
Office 1 : Address : swagat Corner Building,
Near narayani Dham temple,Katraj
Pune
Branches Available : Dhayari Pune, Baramati, Solapur
Opinion Mining & Sentiment Analysis Based on Natural
Language Processing
Abstract
The main goal of social network analysis is the study of structural properties of networks. Structural
analysis of the social network investigates the properties of individual vertices and the global properties
of the network as a whole. It answers two basic classes of questions about the network: what is the
structural position of any given individual node and what can be said about groups forming within the
network. The main measurement of a node’s social power is centrality, which allows to determine
node’s relative and absolute importance in the network. There are several methods to determine node’s
centrality, such as the degree centrality the betweenness centrality or the closeness centrality. The
proposed system will able to mine users’ intent from comments. Irrelevant comments removal will
increase opinion mining performance of system. False positive & false negative rates may reduced.
Resistant to fake opinion postings.
Motivation of the Project
- To overcome the fake posting problems in web application.
- It can be used in any of the social website to find positive and negative comments over the products.
- It will reduce the percentage of fake postings.
System Architecture Diagram
2. Ocular Systems Pune Info@ocularsystems.in
Office 1 : Address : swagat Corner Building,
Near narayani Dham temple,Katraj
Pune
Branches Available : Dhayari Pune, Baramati, Solapur
Modules in the project
To develop website to post advertise & accept comments
Offline comment data collection
POS Feature Extraction
NER Features Extraction
N-Gram Extraction
Train Classifier
Comment-Advertise clarity score
Irrelevant Ad removal
Classification in 3 classes: Positive, Negative, Neutral
Graphical result plotting
3. Ocular Systems Pune Info@ocularsystems.in
Office 1 : Address : swagat Corner Building,
Near narayani Dham temple,Katraj
Pune
Branches Available : Dhayari Pune, Baramati, Solapur
POS & NER Tagger
POS Tagger: POS-tags can be used in extraction of words of a specific word class (all finite verbs,
all nouns, etc.), to decide which word class a word belongs to in a given position (She flies =
verb, the flies = noun), or to group word classes into syntagmata.
Used as important feature set in NLP.
Built by Stanford university
N-Gram Extraction
N-Gram is noting but the collection of keywords appeared in a statement.
For Example:
One Gram: prime
Bi-Gram: Prime Minister
Tri-Gram: Prime Minister India
Along with N-Grams their frequency counts are used as feature set to NLP Classifier.
Porter Stemmer Algorithm
Used for suffix removal.
Plays important role in frequency count mining.
Example: Suppose we want to extract frequency count of “play” in paragraph. Then we simply
use string comparison.
It would be problematic if paragraph contains “plays, played, playing, etc.”
From keywords like “Red”, “ed” should not be removed.
4. Ocular Systems Pune Info@ocularsystems.in
Office 1 : Address : swagat Corner Building,
Near narayani Dham temple,Katraj
Pune
Branches Available : Dhayari Pune, Baramati, Solapur
For this reason porter has designed algorithm for stemming.
Stop Word Removal
All stop words, for example, common words, such as a and the, are removed from multiple word
queries to increase performance of data mining systems.
Steps:
Create database of stopwords
Remove each occurance of stop word which is present in our database.
Naïve Bayes classifier
Naïve Bayes algorithm is probabilistic algorithm
Algorithm takes input in ARFF (Attribute relationship file format) & gives probability for each
class.
If comment says:
“I purchased this product, it seems fake, performance of the gadget is poor”
Classifier would suggest probabilities:
Positive: 2%, Negative 78%, Neutral: 20%
Though algorithm is based on probabilities for prediction, called as NP-Hard problem.
So, we will choose highest probability results to calculate summary.
Software requirement
Operating System : Windows
5. Ocular Systems Pune Info@ocularsystems.in
Office 1 : Address : swagat Corner Building,
Near narayani Dham temple,Katraj
Pune
Branches Available : Dhayari Pune, Baramati, Solapur
Technology : Java and J2EE
Web Technologies : Html, JavaScript, CSS
IDE : Netbeans 7.4
Web Server : Apache Tomcat 6.0
Database : My SQL
Java Version : J2SDK1.6/7
Conclusion
The proposed system will able to mine users’ intent from comments.
Irrelevant comments removal will increase opinion mining performance of system.
False positive & false negative rates may reduced.
Resistant to fake opinion postings.