bhuratna / sentimental-analysis-of-twitter-data-using-machine-learning

Sentimental-Analysis-of-twitter-Data-using-Machine-Learning

ABSTRACT

As the social networks are evolving these days, sentiment analysis and opinion mining are growing area of study. The usage of social networks by people has paved way for analyzing the content and study on people views on products, specific events, social causes, politics etc. The research on these text data is studied by researchers for various purposes such as sentiment analysis, opinion mining, recommendations, etc. In this study, we have proposed analyzing Twitter data and sentiment analysis of data using different machine learning algorithm. The proposed technique provides comparison of performance of machine learning algorithm and feature extraction models namely Count Vectorization models. The experimental results are conducted using feature extraction model and applied machine learning algorithms namely Support Vector Machine (SVM), Random Forest, Naive Bayes, XGBoost, and Decision Tree.

SYSTEM REQUIREMENTS AND SPECIFICATION

Introduction

Machine learning is a rapidly growing dynamic area of research in these days. The machine can able to think like human using these advancements. The recent research in machine learning promises the improvement and precision of sentiment analysis problems. There are many types of machine learning techniques are available, whereas we have taken three types to implement for sentimental analysis.

Requirement Analysis

Software Requirement Specification (SRS) is done as initial step in the software developing activity. When the system becomes complex, the aim of entire system cannot be achieved precise, hence there is a need for requirement management. The software applications are done for client’s requirement. The SRS represents the requirement of client’s view as a formal document. Requirement specification includes specifying the requirement of developing technology, tools used, and checking the specifications are represented during this activity. Generating the SRS document is main aim of this phase. This phase is terminated with a preparation of valid SRS document. The main objective of the Software Requirement Specification is to minimize the communication gap between the client’s requirement and application developers. This is the medium though which the client or user requirements are accurately documented. It forms the basis of software development.
FUNCTIONAL REQUIREMENTS The proposed application should be able to analysis and identify sentiment of taken dataset as positive or negative or neutral. Sentiment analysis is performed by support vector machine (SVM) and Logistic regression, and Random forest techniques.

Data collection

The data collection process involves the selection of quality data for sentimental analysis. Here we taken twitter dataset from extracted from twitter.com. The dataset considered is training set and test set separately. This dataset is considered further for data analysis and comprehensive data, interpreting it, and analyzing results with the help of machine learning or statistical techniques.

Data visualization

A huge amount of data, if represented in graphic way is easier to understand and analyze. In general, data visualization is much helpful to analyze the data nature as big picture. Data preprocessing In machine learning, there is highly needed for pre-processing dataset, the aim of preprocessing is to convert raw data into a form that fits machine learning. Structured and clean data allows a data analysts to get more precise results from machine learning model. The technique includes data formatting, cleaning, and sampling.

Dataset splitting

A dataset used for machine learning can be divided into three subset, they are training set, test set, and validation sets. Training set. A data analyst uses a training set to train a model and define its optimal parameters it has to learn from data. Test set. A test set is needed for an evaluation of the trained model and its capability for generalization.

Model training

After a data analyst has preprocessed the collected data and split it into train and test can proceed with model training. This process entails giving the algorithm with training dataset. A machine learning algorithm used to process data and output a model that is able to find or predict a target value in new data with predictive analysis.

Model evaluation and testing

The goal of this step is to develop the simplest model able to formulate a target value fast and well enough. A data scientist can achieve this goal through model tuning. That’s the optimization of model parameters to achieve an algorithm’s best performance.

Non-functional requirements

The following is a list of non-functional requirements. The specific details will need to be defined by internal stakeholders.  Response Time  Availability  Stability  Maintainability  Usability

SYSTEM REQUIREMENTS

The system requirements includes Hardware and Software requirement, which are provided below System Requirements Hardware Requirements Processor : Any Processor above 500 MHz. Ram : 4 GB Hard Disk : 4 GB Input device : Standard Keyboard and Mouse. Output device : VGA and High Resolution Monitor. Software Requirements Operating System : Windows 7 or higher Programming : Python 3.6 and related libraries MODULES The following modules are implemented in our implementation  Data Set Preparation  Data Pre-process  Feature extraction  Machine learning and Comparison