Scribd
Upload
7 views8 pages

Presentation 16

The document outlines a B.Tech mini project focused on developing a Natural Language Processing (NLP) model for sentiment analysis to improve insights and customer relations in e-commerce and streaming. It discusses the challenges of manual analysis, existing work in the field, and the research methodology adopted for the project. Key challenges in NLP, such as biased training and misinterpretation, are also highlighted.

Uploaded by

rohansahu02
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
7 views8 pages

Presentation 16

The document outlines a B.Tech mini project focused on developing a Natural Language Processing (NLP) model for sentiment analysis to improve insights and customer relations in e-commerce and streaming. It discusses the challenges of manual analysis, existing work in the field, and the research methodology adopted for the project. Key challenges in NLP, such as biased training and misinterpretation, are also highlighted.

Uploaded by

rohansahu02
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/ 8

B.TECH.

MINI PROJECT-1(SE391)
REVIEW ANALYSIS AND SENTIMENT LEARNING USING NLP

SUBMITTED BY:
ROHAN SAHU(2K22/SE/139)
SAHIL(2K22/SE/149)
UNDER SUPERVISION OF
DR. SONIKA DAHIYA
ASSISTANT PROFESSOR
DEPARTMENT OF SOFTWARE ENGINEERING
OUTLINE

• Motivation:
Need to develop this model, and usefulness of model
• Introduction
Basic introduction to model and problem statement
• Existing work
Related work done in the past
• Research methodology
The method we adopt to develop this model
MOTIVATION

• E-commerce and Streaming: User content offers insights


into consumer behavior and trends.
• Manual Analysis Challenges: Manual review is time-
consuming and inconsistent.
• NLP and Sentiment Analysis: NLP aids in categorizing
sentiment, but sarcasm and jargon are challenges.
• Project Goal: Build a system using NLP to improve insights
and customer relations
INTRODUCTION
• Challenge with User Data: The
large volume and complexity of
user-generated content make
manual analysis slow and
inaccurate.
• Project Goal: Develop an NLP-
based system to accurately classify
sentiments and interpret user data
for actionable insights
EXISTING WORK

• TextRank: A graph-based algorithm that represents words as


nodes, capturing semantic similarity through edge weights.
• Word2Vec: A method for learning word representations by
capturing semantic relationships in a continuous vector space.
• CNN in Sentiment Analysis: Used by Meena to classify
sentiment polarity in social media data, achieving 95.4%
accuracy.
• RNN for Amazon Reviews: RNN variants combined with word
embeddings to classify sentiment in reviews, achieving 93.75%
accuracy.
RESEARCH • Natural Language Processing (NLP)
is a field of artificial intelligence
METHODOLO focused on enabling computers to
understand, interpret, and generate

GY human language.
CHALLENGES OF NLP

• Biased Training: NLP models can inherit biases from


training data, affecting fairness and accuracy.
• Misinterpretation: Unclear inputs like slang or background
noise can lead to inaccurate results.
• New Vocabulary: NLP systems struggle with evolving
language and new terms.
• Tone of Voice: NLP has difficulty detecting emotional tone
or intent, such as sarcasm.
REFERNCES

• Qi Han, Junfei Guo and Hinrich Schutze, 2013


Combining an SVM Classifier and Character N-gram
Language Models for Sentiment Analysis on Twitter Text
• Peng Chen, Zhongqian Sun , Lidong Bing, Wei Yang, Recurrent Attention
Network on Memory for Aspect Sentiment Analysis(2017).

• Gaurav Meena, Krishna Kumar Mohbey and Ajay Indian 2021,


Categorizing Sentiment Polarities in Social Networks Data Using
Convolutional Neural Network

You might also like