adsm
كلية أبوظبي للإدارة
ABU DHABI SCHOOL OF MANAGEMENT
Sentiment Analysis using Artificial Intelligence for
governmental organization e-services
AM
Supervisor
Dr. Neda Abdelhamid
Master of Science in Business Analytics (2022-2023)
The candidate confirms that the work submitted is their own and
appropriate credit has been given where reference has been made to the
work of others.
1 adsm
كلية أبــوظــبــي للإدارة
ABU DHABI SCHOOL OF MANAGEMENT
Sentiment Analysis using Artificial Intelligence for
governmental organization e-services
Supervisor
Dr. Neda Abdelhamid
A project presented to
ABU DHABI SCHOOL OF MANAGEMENT
In partial fulfilment
of the requirement for the degree of
Master of Science in Business Analytics (2022-2023)
2 AI
ANN
API
CRISP-DM
ᎠᏴ
HTTP
Abbreviations
A sample of abbreviations is given below. The student should important abbreviations
according to their project
Artificial Intelligence
Artificial Neural Network
Application Programming Interface
Cross-Industry Standard Process for Data Mining
Database
Hypertext Transfer Protocol
IoT
Internet of Things
JSON
JavaScript Object Notation
LSTM
Long Short-Term Memory
ML
NLP
NoSQL
PCA
REST
RNN
SaaS
SQL
SVM
TF-IDF
UAE
TAMM
ADDA
NLP
CRISP-DM
NLTK
SVM
Tweepy
Machine Learning
Natural Language Processing
Not Only SQL
Principal Component Analysis
Representational State Transfer
Recurrent Neural Network
Software as a Service
Structured Query Language
Support Vector Machine
Term Frequency-Inverse Document Frequency
United Arab Emirates
AbuDhabi E-Services Government Platform
Abu Dhabi Digital Authority
Natural Language Processing
Cross-Industry Standard Process for Data Mining
Natural Language Toolkit
Support Vector Machines
an open source Python package that gives you a very convenient way to access the
Twitter API with Python
3 699
889aa0
.7
.8
.8
.10
.10
.10
.11
.11
.12
2
.12
.12
22
.12
.13
Table of Contents
Abbreviations
Chapter 1
Motivation
Background
Related work...
Problem Statement..
Approach and methodology
Table of Contents
1. Business Understanding (CRISP-DM Phase 1):
2. Data Understanding (CRISP-DM Phase 2):
3. Data Preparation (CRISP-DM Phase 3):.
4. Modeling (CRISP-DM Phase 4):.
5. Evaluation (CRISP-DM Phase 5):..
WIPLE
6. Deployment (CRISP-DM Phase 6):
Scope and limitations
Target group
Literature Review
Chapter 2 (Sentiment Analysis using Artificial Intelligence for governmental organization e-services).
Introduction.
Introduction to Abu Dhabi E-Services Platform (TAMM)
Sentiment Analysis in E-services:.
Crisp-DM for sentiment analysis....
Tweepy.
Machine learning for sentiment analysis.
Sentiment Analysis using Arabic Language..
Literature Review Table.
Chapter 3 (Methodology).
Methodology
1. Business Understanding: .
2. Data Understanding:.
Tweet Flash
Exploratory Data Analysis (EDA).
3. Data Preparation:
4. Modeling:
Term Frequency inverse document frequency TF-IDF .
4
.14
1567222222222
.19
.20
.20
.23 .23
Overfitting in machine learning.
Sentiment Analysis of Twitter TAMM Tweets: An Approach
Data Acquisition and Normalization
.24
.25
.25
Data Preparation.
Model Training and Validation.
Machine Learning Model Selection
.25
.25
.26
Outcome
Performance Evaluation and Model Optimization.
Chapter 4 (Data Analysis)
Evaluation.....
Apply the trained ML algorithms to our Tweets.
Analyzing the output of our results.
Overall general comparison.
Pairwise Classifier Agreement Heatmap.
Chapter 5 (Results and Findings).
MLP Classifier
Word Cloud Analysis Findings.
Analysis in depth using the context of frequent words..
Chapter 6 (Discussion).
Sentiment analysis for decision making process..
Hypothesis Validation..
H1: Sentiment
H2: CRISP-DM
as a Structured Methodology for Sentiment Analysis.
H3: Precision of Sentiment
Chapter 7 (Conclusion)...
Future work
References
S
AMPLE
.26
.26
.26
.29
.29
.29
.31
.35
.35
.36
.38
.39
.40
.41
Analysis as a Tool for Enhancing E-Service User Experience
.41
.42
Classifications Through Machine Learning Techniques..
.42
.43
44
.45
5/n Student Note:
This is chapter 7.
To be done in 1000 words.
You already did chapter 1-6.
