A system for twitter user list curation

Abstract

With increased adoption of social networking tools, it is becoming more difficult to extract useful information from the mass of data generated daily by users. Curation of content and sources is an important filter in separating the signal from noise. A good set of credible sources often requires painstaking manual curation, which often yields incomplete coverage of a topic. In this demo, we present a recommender system to aid this process, improving the quality and quantity of sources. The system is highly-adaptable to the goals of the curator, enabling some novel uses for curating and monitoring lists of users.

Finding news curators in twitter

“…Overall, we were able to demonstrate that the considered features can be used 5 The section of an article can be extracted from the prefix of the path of the article. For instance, articles under http: //www.bbc.co.uk/news/world-latin-america-21001060 correspond to the section "Latin America" of BBC.…”

“…sports, business, Europe, USA), we also define the number of distinct sections of the crowds s/he belongs to (UserSectionsQ). 5 …”

“…Wagner et al [24] demonstrated that features based on the user lists perform best, compared to contentbased features based on recent tweet/retweets and features based on the profile. Approaches to automatically extend or create these lists exist as well [5]. Nowadays, these lists are also used to filter valuable content for journalists.…”

CICBUAPnlp at SemEval-2016 Task 4-A: Discovering Twitter Polarity using Enhanced Embeddings

“…Kumar et al (Rangarajan Sridhar, 2015) focused on the spelling issues in social media messages, which includes repeated letters, omitted vowels, use of phonetic spellings, substitution of letters with numbers (typically syllables), use of shorthands and user created abbreviations for phrases. In a data-driven approach, Brigadir et al (Brigadir et al, 2014) apply URL filtering combined with standard NLP preprocessing techniques.…”

“…These approaches treat special characters such as ,.! ?# and user mentions as a regular word (Le and Mikolov, 2014;Brigadir et al, 2014). Still, in some works which use embeddings a basic data cleaning process (i.e., stopwords removal, URL filtering, and removal of rare terms) improves the feature representation and, consequently, the performance of the classification task (Yan et al, 2014;Rangarajan Sridhar, 2015;Jiang et al, 2014).…”