Sentiment analysis meets social media – Challenges and solutions of the field in view of the current information sharing context

Balahur, Alexandra; Jacquet, Guillaume

doi:10.1016/j.ipm.2015.05.005

Cited by 25 publications

(10 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As the technology is further refined, standardization of methodology and the establishment of healthcare specific SA methods (either ML algorithms or a medical‐sentiment lexicon) may facilitate the development of further validity regarding the application of this technology to the health care sector 41,42 …”

Section: Discussionmentioning

confidence: 99%

Sentiment analysis of social media posts on pharmacotherapy: A scoping review

Sharma

Whittle

Haghighi

et al. 2020

Pharmacology Res & Perspec

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Sentiment analysis of social media posts on pharmacotherapy: A scoping review

Sharma

Whittle

Haghighi

et al. 2020

Pharmacology Res & Perspec

View full text Add to dashboard Cite

show abstract

“…Instead, the "triggers" of those emotions are written. The challenge is identifying terms that act as the trigger and associating them with an emotional label [13].…”

Section: Processing Natural Languagementioning

confidence: 99%

“…There are also massive negative implications to the character limit imposed by Twitter. The "brevity" of tweets requires users to include non-standard abbreviations, typos, irony, and trending topics called hashtags [13]. Such unconventional and unstructured texts are considered to be 'noise' as natural language processing (NLP) software does not handle such information so well, creating problems for Twitter content analysis [15].…”

Section: Processing Natural Languagementioning

confidence: 99%

The Perils of Using Social Media Data to Predict the Spread of Diseases

Patel¹,

Amrit

Sundaram

2019

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

The data produced by social media engagement is of interest to various organizations and has been used in different applications like marketing, finance and healthcare. Though the potential of mining this data is high, standard data mining processes do not address the peculiarities of social media data. In this paper, we explore the perils of using social media data in predicting the spread of an infectious disease; perils that are mostly related to data quality, textual analysis and location information. We synthesize findings from a literature review and a data mining exercise to develop an adapted data mining process. This process has been designed to minimize the effects of the perils identified and is thus more aligned with the requirements of predicting disease spread using social media data. The process should be useful to data miners and health institutions.

show abstract

“…Some studies, such as the work of Aidan, Kushmerick, and Smyth (2002), classify features of opinionated documents into two categories: those that depend on the query and incorporate relevance and opinion into the learning phase (Saif et al, 2014;Seki, Kino, Sato, & Uehara, 2007), and those that use characteristics independent of the topic and do not incorporate relevance into the learning phase. Furthermore, while some studies use a single classifier like support vector machine (SVM), naive bayes or logistic regression to return opinionated documents, others use multiple different classifiers to compare their impacts on opinion detection (Balahur, 2016;Balahur & Jacquet, 2015;Bauman, Liu, & Tuzhilin, 2016;Fu, Abbasi, Zeng, & Chen, 2012;Lu, Mamoulis, Pitoura, & Tsaparas, 2016;Mullen & Collier, 2004;Pang & Lee, 2004;Riloff & Wiebe, 2003;Seki et al, 2007;Tu, Cheung, Mamoulis, Yang, & Lu, 2016). Finally, some pre-existing approaches use internal collections built directly from the collection to be analyzed for collections training, while others use external collections built from independent collections of the analyzed collection (Aidan et al, 2002;Baccianella, Esuli, & Sebastiani, 2010;Bifet & Frank, 2010;Pak & Paroubek, 2010;Seki et al, 2007).…”

Section: Related Workmentioning

confidence: 99%

Using language models to improve opinion detection

Belbachir

2018

Information Processing & Management

View full text Add to dashboard Cite

A B S T R A C T :Opinion mining is one of the most important research tasks in the information retrieval research community. With the huge volume of opinionated data available on the Web, approaches must be developed to differentiate opinion from fact. In this paper, we present a lexicon-based approach for opinion retrieval. Generally, opinion retrieval consists of two stages: relevance to the query and opinion detection. In our work, we focus on the second state which itself focusses on de-tecting opinionated documents . We compare the document to be analyzed with opinionated sources that contain subjective information. We hypothesize that a document with a strong si-milarity to opinionated sources is more likely to be opinionated itself. Typical lexicon-based approaches treat and choose their opinion sources according to their test collection, then cal-culate the opinion score based on the frequency of subjective terms in the document. In our work, we use different open opinion collections without any specific treatment and consider them as a reference collection. We then use language models to determine opinion scores. The analysis document and reference collection are represented by different language models (i.e., Dirichlet, Jelinek-Mercer and two-stage models). These language models are generally used in information retrieval to represent the relationship between documents and queries. However, in our study, we modify these language models to represent opinionated documents. We carry out several ex-periments using Text REtrieval Conference (TREC) Blogs 06 as our analysis collection and Internet Movie Data Bases (IMDB), Multi-Perspective Question Answering (MPQA) and CHESLY as our reference collection. To improve opinion detection, we study the impact of using different language models to represent the document and reference collection alongside different com-binations of opinion and retrieval scores. We then use this data to deduce the best opinion de-tection models. Using the best models, our approach improves on the best baseline of TREC Blog (baseline4) by 30%.

show abstract

Sentiment analysis meets social media – Challenges and solutions of the field in view of the current information sharing context

Cited by 25 publications

References 3 publications

Sentiment analysis of social media posts on pharmacotherapy: A scoping review

Sentiment analysis of social media posts on pharmacotherapy: A scoping review

The Perils of Using Social Media Data to Predict the Spread of Diseases

Using language models to improve opinion detection

Contact Info

Product

Resources

About