Search

"Polarity detection" is a technique used in the field of sentiment analysis in texts, a subset of natural language processing (NLP), which determines the emotional polarity of a text, i.e., whether it expresses a positive, negative, or neutral opinion. Despite the impact registered in the field of Natural Language Processing by Deep Learning techniques and, in particular, by Transformers, these approaches come not without downsides, generally related to low-sources languages and domains. In this article, a lexicon-based approach is applied to extract polarity from medical reviews written in Italian from an Italian portal. Specifically, Vader-IT, an adaptation of the popular VADER sentiment analysis tool tailored for the Italian language, is employed to predict the overall sentiment expressed in 5,491 Italian-language reviews about several Italian hospital departments specialized in heart-related diseases. The reviews are accessible on the Qsalute website urlhttps://www.qsalute.it/. The results show a micro averaged F1 −score = 90.8% and a micro averaged Jaccard −score = 83%. [ABSTRACT FROM AUTHOR]

During the pandemic, rapid growth in online learning and educational platforms have been seen. However, the major problem faced by the learners is that they were missing the student-teacher interaction as well as peer-to-peer interaction. To cover up this gap, features like forums, live chat, etc. were introduced. In this project, we will be building a live chat server using Node.js, Express. js, and socket. io. One thing that is observed in these online learning platforms is that the usersmay not always consider the sentiments and feelings of other users present on the live server. To prevent the use of slang, vulgar language, and inappropriate words, we would be using ananalysis bot that will scan for such inappropriate words and if found, the message will be highlighted and will not be uploaded to the server unless corrected. This helps to maintain the quality of chats on the website considering that the website is meant to be a learning platform where users from various backgrounds, age groups, and nationalities will be present. Sentiment analysis, Data Mining, and Natural Language Processing will be used toanalyze the message. The degree of the in appropriateness of the text message will be calculated based on the classification of words done using the Naive-Bayes Classifier. [ABSTRACT FROM AUTHOR]

Twitter is one of the most important social media platforms that is used to keep oneself updated about any political or any news. Twitter generally has a word limit, which also limits the words to be detected in the tweets. Twitter has developed into one of the primary platforms for political discourse and public policy discussions. This paper proposes a Twitter sentiment analyser for political tweets using Machine Learning (ML) and Natural Language Processing (NLP). This paper envisions a model which performs sentiment analysis on tweets discussing a particular political or policy topic from a particular date and performs sentiment analysis on that data. The system is trained with a dataset of 160000 tweets commenting on Indian politics. The system uses NLP for pre-processing and feature extraction. Classification is performed using 3 classifiers namely Random forest, Support Vector Machine (SVM) and Bernoulli Naive Bayes. The highest accuracy obtained was 88.36% using SVM classifier. [ABSTRACT FROM AUTHOR]

Text preprocessing is a crucial stage in natural language processing (NLP). In traditional machine learning algorithms, text preprocessing is often done by cleaning the text, such as removing all stopwords, emojis, symbols. Then proceed with standardization processes such as lower casing, stemming, lemmatization. Even though it often helps the training process with minimal resources and shorter time, this method often eliminates particular meanings, such as the meaning of a sentiment. This research uses a state-of-the-art approach in sentiment analysis with deep learning using word embedding, which is proven to outperform the machine learning approach. We collect datasets through social media, Twitter, and YouTube. Then we test them with various text preprocessing techniques that are most optimal for Indonesian sentiment analysis. The results show that the dataset treated with emoji conversion, multi-word grouping on negated stopwords, slang word conversions, abbreviations, and typo correction can significantly increase the model's accuracy. Meanwhile, the stemming process on the dataset does not help increase the accuracy of the sentiment analysis model. [ABSTRACT FROM AUTHOR]

Social media has become a platform to express views and opinions by people. The posts on social media can be used for opinion mining and the analytics serves as a reference for corporate vision, planning, and implementation. Natural Language Processing is a great tool for social network data analytics. In this paper, sentiment analysis of twitter data is explored. The FLAIR framework is used for sentiment analysis using Named Entity Recognition (NER). The analysis is performed on parameters like frequency, average polarity, total likes and total retweets. The results for the polarity of expressions for the Corona Virus (COVID19) outbreak and its impact is analysed and results are presented here. [ABSTRACT FROM AUTHOR]

Due to the increase in the popularity of the ecommerce portals, the volume of information about products sold online is becoming overwhelmingly abundant in such portals and social networking sites. Buyers usually post their insights about a product in the form of opinions about a product. Extraction of such opinions expressed implicitly or explicitly in a text using Natural Language Processing techniques is referred to as sentiment analysis or opinion mining. Extracting such rationale could give better insights to a manufacturer of a product to make managerial decisions and perform fault diagnosis and can perform potential improvements to a product. Analysis of sentiments of the product reviews can only tell us whether a particular product is liked by the customer or not. However, research works to extract the rationale behind such opinions is scanty. We propose to leverage a deep neural model to identify the rationale for the opinions expressed by customers. [ABSTRACT FROM AUTHOR]

Among the tasks of natural language processing, at present, undoubtedly, the task of analyzing the tonality of data, namely the task of text processing, is an urgent task. In our daily activities, we encounter the most commonly used sentiment analysis applications that allow developers to improve technological solutions and products. For example, by identifying the emotional coloring of a dialogue, quickly determining the nature of a film review, or recognizing the rating of a review about a product, the developed recommendation system helps the user find the most appropriate solution. In present paper, we consider a mood analysis model for both text messages and audio recordings, which classifies an array of input data into two categories of emotions (positive and negative). More than 6000 data characteristics were extracted to execute the model, the source of which was various audio data from the Python openSMILE library. Using the technique of correlation between all the detected properties of audio tracks, it was possible to identify the most sensitive properties most accurately determined by the emotional coloring of audio messages. In addition, since the best sound characteristics are selected, the correlation threshold value was assumed to be 0.9, which ensures that the model will not be retrained. Another Python library, Sklearn, allows you to process text data. It is important to note that both audio and text data are present in enclosures that contain text messages for transcribing audio messages. The developed algorithm consists of several stages. First, the initial data is preprocessed, and stop words, additional spaces, etc. are deleted. Next, training takes place on pre-prepared data. The final step is to run the model on test data. For training, verification and testing, a multimodal open data set Multimodal EmotionLines Dataset (MELD) is used, consisting of more than 1400 dialogues and 13,000 models of utterances from the famous sitcom "Friends". Note that the data arrays are preliminarily divided into three categories of emotional coloring, such as positive, negative and neutral. The results of the computational experiment showed that the model created on two types of data, both text and audio versions, allows for better quality of binary sentiment analysis compared to separate processing of text data or audio data only. [ABSTRACT FROM AUTHOR]

Sentiment analysis is computational research of opinion and emotion that covers textually. This research aims to investigate the Natural Language Processing (NLP) technology in analyzing the sentiment from social media with the Indonesian language. From many methods applied in sentiment analysis, the popular one is Deep Learning (DL). Therefore, this study was conducted to increase sentiment analysis accuracy by knowing the effectiveness of the combination DL algorithm, among others the Convolutional Neural Network (CNN) algorithm and the Long Short Time Memory (LSTM) algorithm. Combining the CNN algorithm and the LSTM algorithm was carried out to process sentiment analysis with tweet data from Twitter. The pre-processing and weighting phase was done using Word2Vec. The result of the experiment with 1295 tweet data showed that the average accuracy of the CNN-LSTM is 71.19% with the highest accuracy of 75.04%. This result indicated that the CNN-LSTM algorithm has the highest accuracy results compared to the basic CNN algorithm and LSTM algorithm. [ABSTRACT FROM AUTHOR]

Over the past few ages, Heart Disease (HD) prediction became a burning and influencing topic of both medical research and scientific field of Machine Learning (ML) and Natural Language Processing (NLP). Early findings will improve the mortality rate of the people and which will aid in reducing the deaths caused due to heart disease. Sentiment Analysis (SA) plays a major role in examining, studying and understanding the heart disease data. Mainly it deals with identifying the disease in early stage by the use of ML techniques with the data available on the internet. [ABSTRACT FROM AUTHOR]

Sentiment analysis is computational research of opinion and emotion that covers textually. This research aims to investigate the Natural Language Processing (NLP) technology in analyzing the sentiment from social media with the Indonesian language. From many methods applied in sentiment analysis, the popular one is Deep Learning (DL). Therefore, this study was conducted to increase sentiment analysis accuracy by knowing the effectiveness of the combination DL algorithm, among others the Convolutional Neural Network (CNN) algorithm and the Long Short Time Memory (LSTM) algorithm. Combining the CNN algorithm and the LSTM algorithm was carried out to process sentiment analysis with tweet data from Twitter. The pre-processing and weighting phase was done using Word2Vec. The result of the experiment with 1295 tweet data showed that the average accuracy of the CNN-LSTM is 71.19% with the highest accuracy of 75.04%. This result indicated that the CNN-LSTM algorithm has the highest accuracy results compared to the basic CNN algorithm and LSTM algorithm. [ABSTRACT FROM AUTHOR]

People's perspectives about products, services, events and other persons are widely available on social media and they represent an important feedback for organizations and governments. In the last decade, with the increased amount of online posted opinions, sentiment analysis (SA) started gaining higher attention from researchers. The aim of SA, an application of natural language processing, is to analyze sentiments within text, and in the case of this work is to extract sentiments associated with positive or negative polarities. Arabic language is of high interest as it is one of the widely used languages. Unfortunately, Arabic SA has many challenges due the special characteristics of the Arabic language which limited the number of works in this area and those that concern the Iraqi dialect in particular. This paper proposes some enhanced classifiers based on hybrid machine learning algorithms combined with two proposed preprocessing steps. These preprocessing steps along with the proposed hybrid classifiers enhance the classification accuracy for the Iraqi dialect reviews which are collected from online movie reviews. The results show up to 12.5% enhancement in the classification performance as compared to the conventional preprocessing and classifiers. [ABSTRACT FROM AUTHOR]

Text mining is one of Artificial Intelligent (AI) technologies that uses Natural Language Processing (NLP) for extracting data from structured text. Sentiment analysis (also known as opinion mining) is one of text mining techniques, which is the computational study of personals' behaviours and opinions. The importance of sentiment analysis increases extensively with the enormous growth of the internet sites such as e-commerce, blogs, social media sites, etc. The aim of the survey is to give full view about the most important sentiment analysis techniques. The paper also represents a major contribution through categorizing the SA techniques in an easy and helpful manner for any researcher through classifying SA based on three essential approaches; machine learning-based, lexicon-based, and hybrid approach. Each approach may contain sub-divisions which in turn represent the techniques of SA classification. [ABSTRACT FROM AUTHOR]

Depression is one of the major causes of increased number of cases of mental illness and suicides all over the globe now a days. A person suffering from anxiety, sadness, depression or suicidal thoughts finds it easier to express his emotions on social media platforms. Thus, the messages or content shared by a person on social media platforms is the best way to detect the mental condition of a person by analysing these messages. Today's situation of pandemic covid19 also increased the cases of depression. In this paper we have used Natural Language processing techniques and deep learning methods to create a model to predict such mental conditions like depression. The model created using LSTM-CNN gives better accuracy of 97% when compared to the other base models of logistic regression, Naïve Bayes, Random Forest and Decision Tree. [ABSTRACT FROM AUTHOR]

Rapid improvement of internet, users express their feelings and views on internet. With the rapid development of social media, different users share valuable location on changing reviews, ideas on different services. Opinion mining is the growing area, which describes different user's feedback about any concept, relates to social media applications from big data sources. Analysis of Sentiment (SA) is the main advantage of Natural Language Processing (NLP) have been interesting concept from past decade years. Generally sentiment/opinion be the classifying the basic representation of text positive, negative and neutral expressions of different users. Traditionally different calculation methods/approaches/algorithms available for mining opinion of different users in social networks. Before applying any algorithm to explore opinion mining there are basic steps available i.e. pre-processing, polarity detection is carried out sentiment over document level, sentiment level like different granularities. This paper describes about different calculations and evaluation of social related sentiment analysis challenges discussed in real time fields. [ABSTRACT FROM AUTHOR]

Sentiment analysis is one part of natural language processing. Sentiment analysis can be done by lexicon based, or machine learning based. Sentiment analysis based on machine learning has advantage of dynamism to meet with new language datasets or new vocabulary. Sentiment analysis seeks to understand the sentiments contained in a sentence. A sentence can be positive, neutral or negative, based on its sentiments. A sentence can have positive, neutral or negative sentiments. However, the fact is each sentence does not always have positive, negative or neutral sentiment clearly. We try to develop a sentiment analysis method that can show the sentiment degree of a sentence. Fuzzy sentiment analysis using convolutional neural network are introduced in this paper to produce more accurate sentiment analysis results. Convolutional neural networks are a popular machine learning method for sentiment analysis. The concept of fuzzy sets is used to express the sentiment degree of a sentence. Euclidean distance analysis to determine the proximity of two vectors is used to show that this method is better than the standard method. The method we propose successfully produces a value that indicates the degree of sentiment of a sentence. Comparison of the euclid distance between the results of the standard sentiment analysis and our method shows that the results of the fuzzy sentiment analysis using convolutional neural network have a distance that is relatively close to the true sentiment value. Fuzzy convolutional neural network analysis sentiment is proven to be able to produce better and smoother sentiment analysis results than standard methods. [ABSTRACT FROM AUTHOR]

Sentiment analysis is an opinion mining that analyzes people's emotions and attitudes on a given topic. The field of sentiment analysis is conducted by Natural Language Processing. In this research, a new sentimental method has been proposed, based on semantic analysis of the reviews, incorporated into the word Embedded as a deep learning framework. Also, in this research proposing a semantic classifier based on part of speech technique that used to identify the polarity of reviews as "positive" or "negative". The concept of Deep learning was implemented in the feature extraction step as an embedding word technique, that used to improve the accuracy of classifying. Due to the limitations in recent conventional bag-of-word models and their ability to capture word meaning similarities, that produce highly dimensioned vectors to throw in learn contextual information about words. The dataset of Internet Movie Database movie reviews was concentrated. Multiple tests were carried out, testing and evaluating the effectiveness of the proposed method using dataset. The recorded results showed a better accuracy by the embedding words as a deep learning technique, compared with the other vector representations. [ABSTRACT FROM AUTHOR]

Understanding the criteria (or "aspects") that reviewers use to evaluate games is important to game developers and publishers, since this will give them important input on how to improve their products. Techniques for the extraction of such aspects have been studied by others, albeit not specific to the gaming industry. In this paper we demonstrate an aspect extraction and analysis system specific to computer games. The system extracts game review texts from a list of known websites and automatically extracts candidate aspects from the review text using techniques from natural language processing and sentiment analysis. It then ranks the candidate aspects using the HITS algorithm. To evaluate the correctness of the extracted aspects, we used the system to calculate an overall score for each game by aggregating its highly-rated aspects, weighted by the importance of the respective aspects. The aggregated scores resulted in a ranking of games, which we compared to a known ranking from a popular website - the rankings showed overall consistency, which suggests that the system has extracted valuable aspects from the reviews. Using the extracted aspect, our system also facilitates the analysis of a game, by evaluating how review articles have rated its performance in these extracted aspects. [ABSTRACT FROM AUTHOR]

Development of internet and technology, has major impact and providing new business called e-commerce. Many e-commerce sites that provide convenience in transaction, and consumers can also provide reviews or opinions on products that purchased. These opinions can be used by consumers and producers. Consumers to know the advantages and disadvantages of particular feature of the product. Procuders can analyse own strengths and weaknesses as well as it's competitors products. Many opinions need a method that the reader can know the point of whole opinion. The idea emerged from review summarization that summarizes the overall opinion based on sentiment and features contain. In this study, the domain that become the main focus is about the digital camera. This research consisted of four steps 1) giving the knowledge to the system to recognize the semantic orientation of an opinion 2) indentify the features of product 3) indentify whether the opinion gives a positive or negative 4) summarizing the result. In this research discussed the methods such as Naive Bayes for sentiment classification, and feature extraction algorithm based on Dependencies Analysis' which is one of the tools in Natural Language Processing (NLP) and knowledge based dictionary which is useful for handling implicit features. The end result of research is a summary that contains a bunch of reviews from consumers on the features and sentiment. With proposed method, accuration for sentiment classification giving 81.2% for positive test data, 80.2% for negative test data, and accuration for feature extraction reach 90.3%. [ABSTRACT FROM AUTHOR]