Opinion Mining

Many studies show that unsafe behavior is the main cause of construction accidents. Safety education and training are effective means to minimise people's unsafe behaviors. Apart from traditional face-to-face construction knowledge sharing, social media is a good tool because it is convenient, efficient, and widely used. We applied both social network analysis and sentiment analysis to investigate knowledge sharing on Twitter. Our study is a novel attempt to understand social structure of "construction safety"-related twitter networks and the opinion leaders. We selected and analyzed 6561 tweets of three users' networks on Twitter-"construction safety", "construction health" and "construction accident". We found that three networks had low density and many isolated vertices, which showed that users did not actively interact with each other. The opinion leaders in this study were mostly organizations or government agencies. The top one is "cif_ireland", the Irish construction industry's representative body, the Construction Industry Federation. 3200 Tweets of the top opinion leader were analyzed through graph metrics calculation, cluster analysis, sentiment analysis, and correlation analysis. The opinion leader used Twitter as a medium to disseminate the latest safety news. Thus, we may use Twitter to stimulate people's interest on construction safety topics, share construction safety knowledge, opinions and ideas. Besides, our results showed that sentiment valence had no correlation with number of favorites or retweets. Nevertheless, there was a positive correlation between favorites and retweets.

The World Wide Web produces vast quantities of data in the form of user thoughts, feelings, viewpoints and claims about various social events, goods, brands, and politics, such as social networks, forums, review sites and blogs. User sentiments shared on the site have a huge effect on readers, software suppliers and policymakers. The unstructured type of social media data needs to be analysed and well-structured, and sentiment analysis has received considerable attention for this reason. The study of feelings is referred to as a text organisation that is used in various ways to identify the conveyed mind-set or emotions, such as negative, good, favourable, unfavourable, thumbs up, thumbs down, etc. A lack of adequate labelling data in the field of natural language processing (NLP) is a problem for sentiment analysis. And to solve this problem, since deep learning models are efficient due to their automatic learning capacity, sentiment analysis and deep learning techniques have been combined. This review paper highlights recent research on the implementation of deep learning and machine learning models to solve various sentiment analysis problems, such as classification of emotions, cross-lingual problems, textual and visual analysis and analysis of product reviews, etc.

Abstract: With the increasing use of web platform such as web blogs, wikis, discussion forums, various other types of social media. People began to share their opinion and experience about product or services on World Wide Web. In this study, we introduce an architecture, implementation, and evaluation of a Web blog mining application, called the Blog Miner, which extracts and classifies people’s opinions and emotions (or sentiment) from the contents of weblogs about movie reviews. As per information visualization with the movie's trend of audiences and reviews becomes important. Movie makers are not only want to know their movie's popularity based on the number of audiences but also check their movie’s evaluation from people who see the movies. If they hope that the movie is a box office hit, they should identify the correlations between audiences and reviews. Here, the system lets the user select a movie and then shows the sentiment score results in a graph.
Keywords:  Web blog, Blog mining, Opinion mining, Sentiment analysis, Web Crawler.
Title: “Analysis and Visualization of Total Movie Review System Using Sentiment Analysis”
Author: Mayuri R.Lahamage
International Journal of Computer Science and Information Technology Research
ISSN 2348-120X (online), ISSN 2348-1196 (print)
Research Publish Journals

Statistički pristupi obradi prirodnih jezika tipično zahtevaju značajne količine anotiranih podataka, a često i različite pomoćne jezičke alate, što ograničava njihovu primenu u resursno ograničenim situacijama. U ovoj disertaciji predstavljena je metodologija razvoja statističkih rešenja u semantičkoj obradi prirodnih jezika sa ograničenim resursima. Ovakvi jezici se odlikuju ne samo limitiranim brojem postojećih jezičkih resursa, već i ograničenim mogućnostima za razvoj novih skupova podataka i namenskih alata i algoritama. Predložena metodologija je usredsređena na kratke tekstove zbog njihove rasprostranjenosti u digitalnoj komunikaciji i zbog veće složenosti njihove semantičke obrade.

Metodologija obuhvata sve faze izrade statističkih rešenja, od prikupljanja tekstualnog sadržaja, preko anotacije podataka, do formulisanja, obučavanja i evaluacije modela mašinskog učenja. Njena upotreba je detaljno ilustrovana na dva semantička problema – analizi sentimenta i određivanju semantičke sličnosti. Kao primer jezika sa ograničenim resursima korišćen je srpski jezik, ali se predložena metodologija može primeniti i na druge jezike iz ove kategorije.

Pored opšte metodologije, u doprinose ove disertacije spada razvoj novog, fleksibilnog sistema označavanja sentimenta kratkih tekstova, nove metrike za utvrđivanje ekonomičnosti anotacije, kao i nekoliko novih modela za određivanje semantičke sličnosti kratkih tekstova. Rezultati disertacije uključuju i kreiranje prvih javno dostupnih anotiranih skupova podataka za probleme analize sentimenta i određivanja semantičke sličnosti kratkih tekstova na srpskom jeziku, razvoj i evaluaciju većeg broja modela na ovim problemima, i prvu komparativnu evaluaciju većeg broja alata za morfološku normalizaciju na kratkim tekstovima na srpskom jeziku.

/

Statistical approaches to natural language processing typically require considerable amounts of labeled data, and often various auxiliary language tools as well, limiting their applicability in resource-limited settings. This thesis presents a methodology for developing statistical solutions in the semantic processing of natural languages with limited resources. In these languages, not only are existing language resources limited, but so are the capabilities for developing new datasets and dedicated tools and algorithms. The proposed methodology focuses on short texts due to their prevalence in digital communication, as well as the greater complexity regarding their semantic processing.

The methodology encompasses all phases in the creation of statistical solutions, from the collection of textual content, to data annotation, to the formulation, training, and evaluation of machine learning models. Its use is illustrated in detail on two semantic tasks – sentiment analysis and semantic textual similarity. The Serbian language is utilized as an example of a language with limited resources, but the proposed methodology can also be applied to other languages in this category.

In addition to the general methodology, the contributions of this thesis consist of the development of a new, flexible short-text sentiment annotation system, a new annotation cost-effectiveness metric, as well as several new semantic textual similarity models. The thesis results also include the creation of the first publicly available annotated datasets of short texts in Serbian for the tasks of sentiment analysis and semantic textual similarity, the development and evaluation of numerous models on these tasks, and the first comparative evaluation of multiple morphological normalization tools on short texts in Serbian.

Social media continues to gain increased presence and importance in society. Public and private opinions about a wide variety of subjects are expressed and spread continually via numerous social media, with Twitter being among the most timely. The ability to quantify and evaluate society’s perceptions is increasingly critical to the success in the marketplace. Sentiment analysis is an approach that can be used to computationally measure perceptions regarding topics based on selected, textual source material. This paper reports on the design of a sentiment analysis tracking algorithm and its implementation in the form of a webbased application that can quantify sentiment contained in Twitter feeds and track the sentiment as it changes over time. Results are provided that evaluate the tool for use in performing tracking of sentiment analysis as perceptions change over time.

This paper explains different methods for sentiment analysis and showcases an efficient methodology. It also highlights the importance the product reviews are of utmost importance for the buyers to decide based on their concerns regarding product's various aspects for example a monitor, processor speed, memory etc. Hence this sentiment analysis of product review provides nearly accurate statistics regarding a product, providing an ease to the customers for analyzing the product and zero down his/her search for an online product. The key focus here is efficient feature extraction, polarity classification thereby summarizing positive and negative or neutral polarity. The proposed work is able to collect information from various sites and perform a sentiment analysis of a user reviews based on that information to rank a product. Also these reviews suffer from spammed reviews from unauthenticated users. So to avoid this confusion and make this review system more transparent and user friendly we propose a technique to extract feature based opinion from a diverse pool of reviews and processing it further to segregate it with respect to the aspects.

Social media has opened new avenues and opportunities for financial banking institutions to improve the quality of their products and services and to understand and to adapt to their cus-tomers' needs. By directly analyzing the feedback of its customers, financial banking institutions can provide personalized products and services tailored to their customer needs. This paper presents a research framework for creation of a financial banking dataset in order to be used for Sentiment Classification using various Machine Learning methods and techniques. The dataset contains 2234 financial banking comments from Romanian financial banking social media collected via web scraping technique.

From outrage at corporations to excitement about innovations, marketplace sentiments are powerful forces in consumer culture that transform markets. This article develops a preliminary theory of marketplace sentiments. Defined as collectively shared emotional dispositions, sentiments can be grouped into three function-based categories: contempt for villains, concern for victims, and celebration of heroes. Marketplace actors such as activists, brands, and consumers have a variety of motives and methods for producing and reproducing sentiments. Activists plant, amplify, and hyper-perform sentiments to recruit consumers and discipline institutions. Brands carefully select, calibrate, and broadcast sentiments to entertain consumers and promote products. Consumers learn, experience, and communicate sentiments to commune and individuate in society. The emergent theory of marketplace sentiments (1) advances a sociocultural perspective on consumer emotion, (2) elevates the theoretical significance of emotional observations in cultural studies, (3) offers a sentiment-based understanding of the power of ideology, (4) indicates how activist sentiments can paradoxically benefit from brand cooptation, and (5) calls for human input in big data sentiment analysis. More broadly, the article proposes that cultures are systems of discourses, sentiments, and practices wherein discourses legitimize sentiments and practices, sentiments energize discourses and practices, and practices materialize discourses and sentiments.

Exploiting sentiment relations to capture opinion targets has recently caught the interest of many researchers. In many cases target entities are themselves part of the sentiment lexicon creating a loop from which it is difficult to infer the overall sentiment to the target entities. In the present work we propose to detect opinion targets by extracting syntactic patterns from short-texts. Experiments show that our method was able to successfully extract 1,879 opinion targets from a total of 2,052 opinion targets. Furthermore, the proposed method obtains comparable results to SemEval 2015 opinion target models in which we observed the syntactic structure relation that exists between sentiment words and their target.

Günümüzde yaygın olarak kullanılan sosyal ağlar, bloglar ve e-ticaret sitelerinde yer alan müşterin tükettikleri ürün ve hizmetler ait yorumlar, diğer müşteriler ve işletmeler için önemli bir bilgi kaynağı haline gelmiştir. Literatürde bu yorumlardan, UGC (User Generated Content) kullanıcı tarafından üretilen içerik veya eWOM (electronic word-of-mouth) elektronik ortamda ağızdan ağıza iletişim olarak bahsedilmektedir.
Araştırmanın amacı; Antalya bölgesinde, helal konseptli otel işletmelerine yönelik müşterileri şikâyetlerinin değerlendirilmesidir. Araştırma örneklemi, Antalya bölgesinde helal konsept çalışan ve helalbooking.com sitesinde yer alan, 5 yıldızlı 7 otel işletmesine ait müşteri şikayetlerini içeren İngilizce metinlerden oluşmaktadır. Araştırma amacı kapsamında otomatik toplanan içerikler metin madenciliği yöntemlerinden fikir madenciliği (bakış tabanlı duygu analizi) tekniği kullanılarak analiz edilmiştir. Rapidminer programında geliştirilen uygulamanın ilk aşamasında müşteri şikâyetlerine ait metinler 15 kategoride özetlenmiştir. İkinci aşamada, müşterilerin negatif bakış açılarına neden olan olay ve varlıklar belirlenmiştir. Araştırma bulgularına göre; Müşterilerin rahatsız olmalarına neden olan en önemli konu başlıkların sessizlik ve otel çalışanlarının (personel) olduğu görülmüştür.  Otellerdeki sessizliği-sakinliğini bozan en önemli unsurların yüksek ses ve müziğin neden olduğu anlaşılmıştır. Ayrıca müşterilerin otel personeli ile iletişim sorunu yaşadığı ve bunun nedeninde yeterli düzeyde İngilizce konuşamamaları olduğu ortaya çıkmıştır.

Abstract
Today's widely used social networks, blogs and e-commerce sites of the customer consumed products and services, comments, and other customers and businesses has become an important source of information. In the literature, these comments are referred to as UGC (User generated Content) or eWOM (electronic word-of-mouth). Purpose of the research; It is the evaluation of the complaints of the customers in the halal concept in the Antalya region. The research sample includes of English texts consisting of 5 star 7 Hotel business, located on the helalbooking.com site of halal concept working in Antalya region. The contents of the study were analyzed by using text mining methods by using opinion mining (aspect-based sentiment analysis) technique. In the first phase of the application developed in the Rapidminer program, the texts of customer complaints are summarized in 15 categories. In the second stage, the events and assets that lead to negative perspectives were determined. According to the research findings; the most important issues that cause customers to be disturbed were the silence and the hotel staff. It is understood that the most important elements that disrupt the silence and calm in hotels are caused by loud sound and music. In addition, customers have experienced communication problems with hotel staff, and the reason for this is that they do not speak enough English.

Sentiment analysis has been increasingly popular in the present digital era which attempts to analyse the consumer reviews acquired from websites, blogs and social media platforms. The rich textual information contained data sources, thus understood as consumers’ reviews are very important to any particular domain as businesses are able to improve themselves in several aspects. This paper focuses on empirical research on sentiment analysis or opinion mining in the healthcare domain. With the careful analysis of all the relevant techniques, the sentiment analysis has secured the leading position in making vital business decisions. It outlines crucial topics pertaining to the sentiment analysis such as the motivation for using sentiment analysis, vital sentiment analysis techniques, new opportunities produced by the analysis, the challenges, and pertinent future directions. It also discusses the relevant data mining techniques and machine learning algorithms involved in the sentiment analysis in the healthcare domain. The study concludes by discussing the main future emphasis of the sentiment analysis in-terms of processing the medical documents to have a better understanding of the medical service consumers.

Nowadays, the social networking is huge and widespread among the world. Because people post, share their opinions, own feelings and own comments for the post and rates the products, photos and videos. Likewise the LINKEDIN which is a Professional website will help the people to post comments, like and share the post, and also help the people search about the job. There is lot of opinions for the single post, but people want the common opinions about the post to clearly know about the job/post. In that situation, sentiment analysis will help the people to know the common opinions about the post/job are positive or negative or neutral. In this paper we have analyzed the people's opinion in the text format about the post whether it is positive or negative or neutral using the NLP (Natural Language Processing)[1], Linguistic Process with Python Coding.

Traditionally, earthquake impact assessments have been made via fieldwork by non-governmental organisations (NGO's) sponsored data collection; however, this approach is time-consuming, expensive and often limited. Recently, social media (SM) has become a valuable tool for quickly collecting large amounts of first-hand data after a disaster and shows great potential for decision-making. Nevertheless, extracting meaningful information from SM is an ongoing area of research. This paper tests the accuracy of the pretrained sentiment analysis (SA) model developed by the no-code machine learning platform MonkeyLearn using the text data related to the emergency response and early recovery phase of the three major earthquakes that struck Albania on the 26th November 2019. These events caused 51 deaths, 3000 injuries and extensive damage. We obtained 695 tweets with the hashtags: #Albania #AlbanianEarthquake, and #albanianearthquake from the 26th November 2019 to the 3rd February 2020. We used these data to test the accuracy of the pre-trained SA classification model developed by MonkeyLearn to identify polarity in text data. This test explores the feasibility to automate the classification process to extract meaningful information from text data from SM in real-time in the future. We tested the no-code machine learning platform's performance using a confusion matrix. We obtained an overall accuracy (ACC) of 63% and a misclassification rate of 37%. We conclude that the ACC of the unsupervised classification is sufficient for a preliminary assessment, but further research is needed to determine if the accuracy is improved by customising the training model of the machine learning platform.

On today's fast spreading use of social media, many websites have offer reviews of items like books, cars, mobiles, movies etc. They describe the product in some detail and evaluate them as good/bad, preferred/not preferred, so it is necessary to categorize these reviews in an automated way. Sentiment analysis is one kind of computational technique of Artificial Intelligence. It is the task of identifying positive and negative opinions, emotions, and evaluations. This article represents Sentiment analysis, it's issues, applications and some of the methods used to evaluate the review using sentiment analysis.

Ever increasing number of internet users has attracted many of the companies on the internet for promoting their product and services. This has led to the development of new age of advertising called online or web advertising. The objective of this chapter is two-fold. First, it introduces concepts involved in
online advertising. Second, it proposes a novel conceptual framework for contextual online advertising which attempts to utilize local context and sentiment for identifying relevant ads. Contextual advertising is an important class of online advertising in which ads are displayed automatically on web pages
based on their content. The proposed framework works in two stages. The first stage retrieves ads for placement. The second stage uses sentiment analysis to filter out ads that do not agree with the sentiments (positive or negative polarity) being expressed in the document. The polarity is identified using
SentiWordNet and context-based heuristics.

Survey research has an important place in marketing evaluations and assessments, and in opinion attitude polling and a wide range of behavioral-social research. It is somewhat surprising that certain inefficient and time consuming procedures in survey research have received but little attention.

This paper reports upon a major and fundamental development in survey-behavioral research. The Questronic project based at the University of Sheffield and its first commercialized product, The Ferranti Market Research Terminal (MRT), focuses upon the procedures of data recording techniques, data preparation and evaluation and, for example, has sought to remove the tedious and costly subsequent encoding and data preparation for input to the computer from field situations. Hopefully, the achievements of the Questronic project to-date will provide an impetus for behavioral research in general, and will be of particular advantage in market, opinion, economic, social and medical-survey research programmes.

Social networks analysis studies the interactions among users when using social media. The content provided by social media is composed of essentially two parts: a network structure of users' links (e.g. followers, friends, etc.) and actual data content exchanged among users (e.g. text, multimedia). Topic modeling and sentiment analysis are two techniques that help extracting meaningful information from large or multiple portions of the text: identifying the topic discussed in a text, and providing a value characterizing an opinion respectively. This extracted information can then be combined to the network structure of users' links for further tasks as predictive analytics, pattern recognition, etc. In this paper we propose a method based on graph databases, topic modelling and sentiment analysis to facilitate pattern extraction within social media texts. We applied our model to Twitter datasets, and were able to extract a series of opinion patterns.

Product review in e-commerce can help prospective
buyers to make decision whether the product that they want
to buy is good or bad and it can help the sellers to get their
consumers feedback. But, there are two problems exist: the
number of product review increases day by day and e-commerce
allows their consumers to write positive negative opinion about
some product features in one review, as known as ”free format”.
Therefore, there should be system which can extract product
feature from review and classify their opinion automatically.
Class Sequential Rule (CSR) method can be implemented in
product feature extraction and Opinion Lexicon method can be
implemented in feature opinion classification. The best F-score of
feature extraction using CSR in free format review is 51.26% and
the best f-score of opinion classification using Opinion Lexicon
in free format review is 35.65%.