2. Related Work
Many of works have been done in the area of public opinion extraction, some works went to find the polarity and others find both polarity and subjectivity. The work has been done for many languages and to serve many of purposes like politics, social services, movie reviews…etc., but unfortunately there is no work has been done for Arabic language. In the following we browsing some of these works:
OSVision Opinion Mining [6] is an automatic system which can extract opinions from the Web. The system uses advanced natural language processing algorithms for extracting opinion, the system is supported by machine learning approaches and knowledge representations which enabled to apply it.
The author mentioned his system, what the objective from this system. But he didn't declare what are the techniques used and how these techniques worked to perform this work, there are no details, and the work method was ambiguous to me, so it is difficult to benefit from this paper.
OPTIMISM [5] an opinion mining system for classification of opinions about related Portuguese political actors. The work passed through several phases as following:
Subscribing the newsfeeds associated to the relevant political actors for collecting the concerned data.
Developing an ontology of political entities which contains the names of the political actors , their variants ( acronyms), and their roles in the political scope.
Create high precision lexico-syntactic rules.
Developing a reference corpus, starting by collecting opinion-rich data from the web site of one of the most popularity Portuguese newspapers.
Manually the training set expressing positive and negative opinions was chosen, and the residual sentences of the commen...
... middle of paper ...
...May 22-24, 2009, pp. 270-272.
[4] J.Froelich, S. Ananyan, D. Olson, The Use of Text Mining to Analyze Public Input, Megaputer Intelligence, 2004.
[5] M. Silva, P. Carvalho, L Sarmento, et al. "The Design of OPTIMISM, an Opinion Mining System for Portuguese Politics", New Trends in Artificial Intelligence: Proceedings of EPIA 2009 - Fourteenth Portuguese Conference on Artificial Intelligence, Universidade de Aveiro, October, 2009, pp. 565-576.
[6] OPINION MINING: Automated Opinion Polls and Product Review Surveys, OSVision.
[7] Y. Almas, K. Ahmad, “A note on extracting ‘sentiments’ in financial news in English, Arabic & Urdu”, The Second Workshop on Computational Approaches to Arabic Script-based Languages” LSA 2007 Linguistic Institute Stanford University, July 21, 2007.
[8] Y. Kuat, M. Saˇsa, " Sentiment Analysis of Movie Review Comments ", 2009.
New York: Random House, 2001. Web. The Web. The Web. http://site.ebrary.com/lib/priority/docDetail.action?docID=10235241>. Buchanan, Albert Russell.
Jean Carletta, “Assessing agreement on classification tasks: The kappa statistic”. Computational Linguistics, MIT Press Cambridge, MA, USA, Vol. 22, No.2, pp. 249–254, 1996.
Healthcare: Sentiment analysis has wide-scale applications in the Healthcare industry. Many patients use internet to post their patient experience in provider facilities. This unbiased feedback from patients is critical for healthcare practices to improve the quality of care. It is not possible for the patient to keep going back to the facility to report post intervention feedback. Extracting patient sentiments from unstructured data in blogs, twitter, Facebook posts help hospitals realize important performance factors like patient satisfaction, staff friendliness, procedure efficiency. Patients also share information regarding their payer experience on the internet. Tweets, posts about insurance benefits, timely service are critical information to the payers to improve their existing services. 94% patients believe hospital’s brand name is important in making a selection. By understanding patient sentiments and taking appropriate action to translate negative feedback into improved care can help a hospital improve its brand image.
Clustering This is un-supervised learning method. Text documents here are unlabelled and inherent patterns in text are revealed through cluster formation. This can also be used as prior step for other text mining methods.
There is a debate between the benefits and potential informational privacy issues in web-data mining. There are large amount of valuable data on the web, and those data can be retrieved easily by using search engine. When web-data mining techniques are applied on these data, we can get a large number of benefits. Web-data mining techniques are appealing to business companies for several reasons [1]. For example, if a company wants to expand its bu...
One would hardly think of politics and data going hand in hand. Well, today political scientists across the globe believe data to be an integral part of political strategies. Computer-automated analysis of blog postings, web traffic analysis post a political speech, tag cloud analysis, social network ‘likes’ and ‘dislikes’ analysis provide a much more educated view of the public sentiment.
Optimistic attitude is a great way to feel better, even during bad times. The interesting question is, whether it can help the optimistic person to live the happy time longer, than his / her pessimistic colleague. The scientists (Maruta, Colligan, Malinchoc, and Offord (2000)) studied this question. They made an experiment: using the data gathered in the mid-1960, they divided the patients in three main groups. The first group was the optimistic, second – mixed, and the last pessimistic. The results were quite unambiguous: for every ten points increase in person’s score on their optimism scale, the risk of early death decreased by nineteen percents. It is a very good result, because, as we can see, the level of optimism is making the life of the peop...
...has so much power. The findings of this research could be used by campaigners in an attempt to swing an election in their favour, creating an unfair bias in parliament and denigrating the ideals of democracy.
There are many studies that examine the direct relationship between news, information and activity online and the subsequent market characteristic. However, I have selected a sample of papers to look at, some of which look at the financial theory behind the stock market, and then several which look at the sentiments which can be extracted from Twitter and online sources and then tested to see if there are any significant relationships present, which could be then use...
Consumers usually express their sentiments on public forums, Social network sites like Twitter. Opinions and feelings are expressed in different way, with different vocabulary, context of writing, usage of short forms and slang, making the data huge and disorganized. Here we extract data from twitter, Manual analysis of sentiment data is virtually impossible. Therefore, python programming language is the best choice using tweepy library .to extract data you must have an account in twitter to use the twitter Application Program Interface (twitter API) which allows the user to reach twitter information as developer. By using twitter streaming API you can extract data from
Sentiment analysis, also called as opinion mining, is the field of study that analyzes people’s opinions, sentiments, evaluations, appraisals, attitudes and emotion towards entities such as products, services or organizations, individuals, issues, topics and their attributes. Sentiment analysis and opinion mining mainly focuses on opinions which express or imply positive, negative or neutral sentiments. Due to the big diversity and size of social media there is a need of automated and real time opinion extraction and mining. Mining online opinion is a form of sentiment analysis that is treated as a difficult text classification task.
Another thing that I learned through this experience is how difficult it can be to code the tone of paragraphs as positive, negative, or neutral. This is a completely subjective task, and it is easy to see how there could be some disagreement in assigning scores between different coders. What one person sees as being a negative paragraph in tone, another person may read the same thing and see it as being neutral. I spent quite a bit of time contemplating how to code each paragraph. Since the accuracy of the results depend on coding the paragraphs with consistency between coders, I used my best judgement and tried to think about how others would code each paragraph as well.
The field of Computational Linguistics is relatively new; however, it contains several sub-areas reflecting practical applications in the field. Machine (or Automatic) Translation (MT) is one of the main components of Computational Linguistics (CL). It can be considered as an independent subject because people who work in this domain are not necessarily experts in the other domains of CL. However, what connects them is the fact that all of these subjects use computers as a tool to deal with human language. Therefore, some people call it Natural Language Processing (NLP). This paper tries to highlight MT as an essential sub-area of CL. The types and approaches of MT will be considered, and limitations discussed.
Mainstream media such as television, radio, newspapers were the primary source of reliable information before the epoch of the internet. However, the situation has changed. The evolution of modern technology in the world today has led to the continuous increase in the methods of practicing journalism. Social and technological advancements have not only improved the pace and content of this field’s practice, but has extended its genre to online or cybernetic journalism. (Project for Excellence in Journalism, 2007). News websites most of which are owned by major media companies and alternative websites with user generated content such as social networking sites and blogs are gaining grounds in the journalism field of practice. (Nel, n.d). One of the chief forces affecting the practice of journalism nowadays is online citizen journalists. Nel (n.d) defines citizen journalism as “individuals playing an active role in the process of collecting, reporting, analysing and disseminating news and information”. He further adds that “citizen journalism is slowly being looked upon as a form of rightful democratic ways of giving hones news, articles, etc, directly by citizens of the world from anywhere.” One of the major researches conducted in the field of citizen journalism, describes the phenomenon as “individuals who intend to publish information online, meant to benefit a community”, and this information is expected to benefit the audience or the wider population in making decisions for the improvement of their community. (Carpenter, 2010.)
There are three reasons for selecting Hasawi for this study. First, there is little previous work related to Hasawi even though it is considered an enormous dialect because it is expanded to other Gulf countries, such as Bahrain, Iran, Iraq, Kuwait, Qatar, and United Arab Emirates. Thus, HD is also spoken as a minor dialect in the previous mentioned countries so that Hasawi is sometimes called Gulf Arabic 'Khaliji'. Secondly, the emergence of a new dialect a few years ago which is Modern Hasawi, a blend of old Hasawi and Najdi, threatens the existence of the original Hasawi in Saudi Arabia in spite of the massive expansion of the dialect to the neighboring countries. Finally, the dialect of Al-Ahsa is seen as a humorous matter among other Saudi dialects because it is hard to understand (Bassiouney, 2010). Probably the cause of such difficulty refers to the sociolinguistic impact of non-Arabian languages, such as Farsi 'Persian' and Turkish. However, it would be proven at the end of this paper that this unattractive dialect has unique acoustic features.