In the last few years use of social networking sites has increased tremendously. People use social media platforms to share their views on almost all subjects. These views are in various forms like, blogs, tweets, Facebook posts, online discussion boards, Instagram posts, etc. Sentiment analysis deals with the process of computationally defining and classifying the views expressed in the comment, post or document. Typically, the aim of sentiment analysis is to find out the customer's attitude towards a product or service. Customers' feedback is vital for businesses, and social media being a powerful platform, can be used to improve and enhance business opportunities if the feedback on social media can be analyzed timely. Therefore, the focus of this paper is to analyze the customer reviews about various restaurants across Karachi - one of the biggest cities of Pakistan. For this research, customer reviews are collected from a very popular Facebook community- the SWOT'S guide to Karachi's restaurants. The contribution of this research is twofold. First, it performs sentiment analysis and classifies each comment as positive, negative. Second, by using text categorization techniques, comments are automatically classified according to feedback about food taste, ambiance, service, and value for money. A manually annotated dataset of around 4000 records was used for training and testing. Several algorithms were used for classification, including Naive Bayes Classifier, Logistic Regression, Support Vector Machine (SVM), and Random Forest. The performance comparison of these algorithms is presented. The best results, that is 95% accuracy, were achieved by using a random forest algorithm.

Sentiment Analysis and Classification of Restaurant Reviews using Machine Learning

Text categorization is an effective activity that can be accomplished using a variety of classification algorithms. In machine learning, the classifier is built by learning the features of categories from a set of preset training data. Similarly, deep learning offers enormous benefits for text classification since they execute highly accurately with lower-level engineering and processing. This paper employs machine and deep learning techniques to classify textual data. Textual data contains much useless information that must be pre-processed. We clean the data, impute missing values, and eliminate the repeated columns. Next, we employ machine learning algorithms: logistic regression, random forest, K-nearest neighbors (KNN), and deep learning algorithms: long short-term memory (LSTM), artificial neural network (ANN), and gated recurrent unit (GRU) for classification. Results reveal that LSTM achieves 92% accuracy outperforming all other model and baseline studies.

/pdf/an-efficient-approach-for-textual-data-classification-using-15iuxk0p.pdf

An efficient approach for textual data classification using deep learning

Public opinion is a public judgment regarding an event or a product. Traditional opinion data mining is often done to get conclusions about how influential an event or product is on society. In the current microblogging era, individual views of events and products can often be found in posts, such as Twitter. Based on the Twitter users’ habits, opinion data can be collected young using the crawling process of events or products. In the event of a pandemic, the use of e-money has increased quite drastically. The use of e-money simplifies the transaction process because everything can be processed only through a smart device. However, the increasing number of e-money users, especially during a pandemic, may not be in line with the increasing positive opinion of e-money. However, some may use e-money only out of necessity. We present classification performance results for the C4.5 decision tree and Naive Bayes classifiers. Lastly, we compare each classifier’s output using the voting-by-majority method to final predict positive or negative opinions about e-money in Indonesia. The proposed method can classify datasets from Twitter posts

C4.5 and Naive Bayes for Sentiment Analysis Indonesian Tweet on E-Money User During Pandemic

Classification of text documents on online media is a big data problem and requires automation. Text classification accuracy can decrease if there are many ambiguous terms between classes. Hadoop Map Reduce is a parallel processing framework for big data that has been widely used for text processing on big data. The study presented text classification using genetic programming by pre-processing text using Hadoop map-reduce and collecting data using web scraping. Genetic programming is used to perform association rule mining (ARM) before text classification to analyze big data patterns. The data used are articles from science-direct with the three keywords. This study aims to perform text classification with ARM-based data pattern analysis and data collection system through web-scraping, pre-processing using map-reduce, and text classification using genetic programming. Through web scraping, data has been collected by reducing duplicates as much as 17718. Map-reduce has tokenized and stopped-word removal with 36639 terms with 5189 unique terms and 31450 common terms. Evaluation of ARM with different amounts of multi-tree data can produce more and longer rules and better support. The multi-tree also produces more specific rules and better ARM performance than a single tree. Text classification evaluation shows that a single tree produces better accuracy (0.7042) than a decision tree (0.6892), and the lowest is a multi-tree(0.6754). The evaluation also shows that the ARM results are not in line with the classification results, where a multi-tree shows the best result (0.3904) from the decision tree (0.3588), and the lowest is a single tree (0.356).

/pdf/text-classification-using-genetic-programming-with-3e2gdvds.pdf

Text Classification Using Genetic Programming with Implementation of Map Reduce and Scraping

In recent years, the popularity of social networking sites has skyrocketed. Social media allows individuals to express their opinions on almost every topic. These opinions take variety of ways, such as tweets, Instagram posts, blogs, Facebook posts, and so on. The process of describing and categorizing the opinions represented in the form of post, document or comment is known as sentiment analysis. The goal of sentiment analysis is typically to determine the attitude of customers toward a good or service. Customer's feedback is critical for organizations and social media that can serve as a promising tool to enhance and improve business opportunities if reviews have been evaluated on social media in a timely manner. As a result, the purpose of this study is to examine online restaurant reviews given by customers. The proposed research makes two contributions. First, it conducts sentiment analysis (SA) and categorizes each and every comment given in the text into five labels i.e.; positive, negative, neutral, highly positive and highly negative. Second, comments are automatically characterized using text categorization techniques based on the feedback from the customers about taste, service, ambiance, food, and value for money. For training and testing, dataset available from Kaggle was used. For classification, numerous algorithms were employed, such as the Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), and Naive Bayes Classifier (NB). This method evaluates the performance of these algorithms and random forest algorithm produced the best results, with highest accuracy.

Feature level Fine Grained Sentiment Analysis for Classifying Online Restaurant Reviews

Text classification is an essential advance in characteristic dialect processing. It very well may be performed utilizing different classification algorithms. Hadoop Map Reduce is widely utilized in text classification to perform classification on colossal measure of text data. However, Map Reduce required a ton of time to perform the tasks thereby increasing latency and since the data is distributed over the cluster it builds time and thus reducing processing speed. Also, Hadoop utilizes long queue of code. Motivated by this, we propose a basic yet compelling machine learning method which uses Naive Bayes classifier for text data. In Machine Learning approach, the classifier is built automatically by learning the properties of categories from a set of predefined training data. Hence, it can process complex furthermore, multi assortmentinformation in dynamic situations. Here we propose a naive bayes classifier which scales directly with number of indicators and data points which can be used for both binary and multi-class classification problems. We implemented the presented schemes using Machine Learning tool. The experimental results demonstrate the performance improvement in the classification technique.

K. V. Ranjitha

Papers

Optimization Scheme for Text Classification Using Machine Learning Naïve Bayes Classifier