Applying sentiment analysis for tweets linking to scientific papers

Bachelorarbeit , 2015 , 66 Seiten , Note: 1,3

BWL - Informationswissenschaften, Informationsmanagement

This work analyzes tweets linking to scientific papers to find out if the tweets are positive, or negative or do not express an opinion. This will inform the meaning of tweets as a measure of impact in the context of altmetrics. The following research questions are examined:

- In how far can sentiment analysis be used to detect positive or negative statements towards scientific papers expressed on Twitter?
- Do tweets linking to scientific papers express positive or negative opinions? How do sentiments differ by academic discipline?
- How do results affect the meaning of tweets to scientific papers as an altmetric indicator?

Inhaltsverzeichnis (Table of Contents)

• Introduction
• Materials and Methods
  • Dataset
    • Bibliographic information of tweeted documents
    • Tweets
  • Sentiment analysis
    • The definition of a sentiment
    • Sentiment analysis
  • Methods
    • Intellectual coding of sentiments
    • Removing Twitter affordances
    • Removing title terms
    • Adapting the lexicon
    • Removing non-English terms
    • Calculating sentiments per discipline
• Results and Discussions
  • The ground truth
    • Sentiment analysis I
    • Sentiment analysis II
    • Sentiment analysis III
    • Sentiment analysis IV
  • Automated analysis of all tweets
  • Results
  • Discipline specific results
• Conclusion

Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)

This work investigates the presence and nature of sentiments expressed in tweets linking to scientific papers. The goal is to understand whether tweets can be used as a reliable indicator of impact in the context of altmetrics, given that they may contain positive, negative, or neutral opinions about the linked papers.

• The use of sentiment analysis to detect positive or negative statements towards scientific papers on Twitter.
• The prevalence of positive, negative, and neutral opinions in tweets linking to scientific papers.
• The influence of academic discipline on the expression of sentiments in tweets.
• The implications of sentiment analysis findings for the interpretation of tweets as an altmetric indicator.
• The importance of considering the value freedom of scientific statements and the potential influence of personal judgments in online communication.

Zusammenfassung der Kapitel (Chapter Summaries)

The introduction defines altmetrics within the context of scholarly communication and reviews relevant literature. It highlights the importance of analyzing sentiments in online discussions about scientific documents, particularly in the era of altmetrics where the number of tweets is often used as a measure of impact.

The materials and methods section details the dataset used (including bibliographic information of tweeted documents and the tweets themselves), the tools employed for sentiment analysis, and the techniques implemented for analyzing sentiments. These techniques include intellectual coding of sentiments, removing Twitter affordances, removing title terms, adapting the lexicon, removing non-English terms, and calculating sentiments per discipline.

The results and discussions section presents and evaluates the results obtained through the methods described in the previous section. This includes an analysis of the ground truth, an automated analysis of all tweets, and a discussion of the results, including discipline-specific findings.

Schlüsselwörter (Keywords)

The primary focus of this work lies in applying sentiment analysis to tweets linking to scientific papers. Key terms and concepts include altmetrics, scholarly communication, sentiment analysis, academic discipline, Twitter, impact, and value freedom. The study investigates the potential influence of personal judgments in online discussions about scientific work, particularly in the context of social media platforms like Twitter.

• Hochschule: Heinrich-Heine-Universität Düsseldorf (Institut für Sprache und Information)
• Note: 1,3
• Autor: Natalie Friedrich (Autor:in)
• Erscheinungsjahr: 2015
• Seiten: 66
• Katalognummer: V312043
• ISBN (eBook): 9783668112704
• ISBN (Buch): 9783668112711
• Sprache: Englisch
• Schlagworte: Altmetrics, Twitter, Impact Factor, Sentiment analysis, Scientific communication