Corpus-based dictionaries for sentiment analysis of specialized vocabularies

Rice, Douglas; Zorn, Christopher

doi:10.1017/psrm.2019.10

Cited by 70 publications

(50 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Rice and Zorn [103] describe general approaches. In one, they suggest that machine learning should approach the process by "classifying or scoring a subset of texts (usually documents) on their sentiment, and then using their linguistic content to train a classifier" [103]. In another approach they suggest "dictionary-based approaches…”

Section: Building a Custom Liwc Dictionarymentioning

confidence: 99%

“…[that] begin with a predefined dictionary of positive and negative words, and then use word counts or other measures of word incidence and frequency to score all the opinions in the data" [103]. Both approaches are highly dependent on having high quality data with comprehensive values based on theory.…”

Section: Building a Custom Liwc Dictionarymentioning

confidence: 99%

“…The validity of a custom dictionary relies heavily on data quality and comprehensiveness in the way the dictionary captures specialized aspects of language use in the relevant texts [100]. Therefore, the current study's DES and SD custom dictionaries were validated with a multi-method approach suggested by Rice and Zorn (2013). The classified words were provided to 4 modeling experts, with over 15 years of simulation experience.…”

Section: Building a Custom Liwc Dictionarymentioning

confidence: 99%

See 2 more Smart Citations

Using LIWC to choose simulation approaches: A feasibility study

McHaney

Tako

Robinson

2018

Decision Support Systems

View full text Add to dashboard Cite

Can language usage help determine which model approach is best suited to provide decision makers with desired insights? This research addresses that question through an investigation of Linguistic Inquiry and Word Count (LIWC), which calculates the presence of more than 80 language dimensions in text samples, and permits construction of custom dictionaries. This article demonstrates use of LIWC to ensure better problem/model fit within the context of selecting a decision support tool. We selected two simulation tools as research instruments to investigate a broader question on the usefulness of LIWC to guide choice of DSS tool. The tools selected were System Dynamics (SD) and Discrete Event Simulation (DES). First, we tested LIWC to analyze practitioners' language use when developing models. LIWC pointed out significant linguistic differences consistent with prior theoretical work, based on model development approach in a number of dimensions. These differences provided a basis for developing a custom dictionary for use on the second part of our study. The second part of the study focused on language used by decision makers in problem statements and used the linguistic clues identified in the first part of the study to ensure problem/model fit. Results indicated problem statements contained linguistic clues related to the type of information desired by problem solvers. The article concludes with a discussion about how LIWC and similar tools can help determine which DSS tools are suited to particular applications.

show abstract

Section: Building a Custom Liwc Dictionarymentioning

confidence: 99%

Section: Building a Custom Liwc Dictionarymentioning

confidence: 99%

Section: Building a Custom Liwc Dictionarymentioning

confidence: 99%

See 1 more Smart Citation

Using LIWC to choose simulation approaches: A feasibility study

McHaney

Tako

Robinson

2018

Decision Support Systems

View full text Add to dashboard Cite

show abstract

“…The ML methods are more useful when the sentences contain words that are more relevant to a particular area/domain, for example, on any medicine/health blog, comments are more word specific which requires a special library for pharmacy area. On the other hand, 'Lexicon'-based methods have a predefined list of words in the dictionary and does not require a training and test dataset (Rice & Zorn, 2013). This method is more useful where the comments are general in nature and also incur lesser cost and time in comparison to ML methods.…”

Section: Lexicon-based Sentiment Analysismentioning

confidence: 99%

Analysing Customers Reactions on Social Media Promotional Campaigns

Saxena

Chaturvedi²,

Rakesh

2018

Paradigm: A Management Research Journal

View full text Add to dashboard Cite

Nowadays, it is a general practice by all major firms to launch social media promotional campaigns/ advertisements while launching their new products; however, no study has discussed the customer's reactions on these social media promotional campaigns. In this study, we have analysed customer reactions on social media promotional campaigns launched by three major mobile phone companies selling their phones in India. The analysis has been done on 18,659 tweets by performing part of speech (POS) categorization-based thematic analysis, Harvard-IV lexicon-based sentiment analysis and social media index-based reach analysis. The results of the study showed that there was a significant difference in customer's reactions on different brands promotional campaigns; ranging from very high social media reach and customer engagement to low social media reach and no customer engagement. Based on its results, the study also proposes major recommendations for marketing managers who are in the process of creating social media promotional campaigns for new products.

show abstract

“…Several research works utilized automatic seed selection using a set of words as initial seed for the dictionary. Turney's [8] and Rice & Zorn [9], built a word of seeds using two (2) human-selected seed words (the word "Poor", "Bagel" as negative and "Excellent", "Love" as positive). In another study, Zagibalov and Carroll [10] also utilized this approach and started with only a single, human-selected seed ("Good").…”

Section: Building Of Sentiment Dictionarymentioning

confidence: 99%

Combining rule-based and bag-of-words for phase-level sentiment analysis of blog comments

Patacsil

2018

IJET

View full text Add to dashboard Cite

Blogs are one of the platforms that express personal opinions, which are intended to create awareness and used as an instrument to establish trust among customers products and services or about a specific topic. A new classifying model was experimented to improve the sentiment classification of the blog comments. This technique combined Rule-Based (RB) and Bag-of-Word (BoW) model to solve major weaknesses of this Bow model in the conduct of Sentiment Analysis (SA) evaluations. The proposed technique was experimented to esti-mate the Philippine Internet customers’ satisfaction related to the quality of the services provided by the ISPs in the Philippines. In addition, automatic word seeding, building of sentiment dictionary utilizing an online dictionary, n-gram, tokenization, stemming and other SA tech-niques were applied to extract useful information from the blog comment dataset. The results of the research showed that the configurations involving BoW-RB + SVM + bi-gram + Porter stemmer achieved a high classification accuracy of 88%.Capturing the contextual meaning of the neighboring words of a given sentimental word provides a significant help to increase the classifi-cation performance of the proposed classifier method.

show abstract

Corpus-based dictionaries for sentiment analysis of specialized vocabularies

Cited by 70 publications

References 31 publications

Using LIWC to choose simulation approaches: A feasibility study

Using LIWC to choose simulation approaches: A feasibility study

Analysing Customers Reactions on Social Media Promotional Campaigns

Combining rule-based and bag-of-words for phase-level sentiment analysis of blog comments

Contact Info

Product

Resources

About