Twitter Topic Fuzzy Fingerprints

Rosa, Hugo; Batista, Fernando; Carvalho, João Paulo

doi:10.1109/fuzz-ieee.2014.6891781

Cited by 30 publications

(30 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We achieved a 4.6% improvement in F1 scores over the results reported by [13]. Then we created a Fuzzy Fingerprints Event library, and adapted the similarity score proposed in [5] to retrieve events in order to improve the results when using all the event types in the ACE 2005 corpus. This paper is organized as follows: Section 2 introduces the related work.…”

Section: Introductionmentioning

confidence: 70%

“…In this work, we propose the use of an adaptation of the Fuzzy Fingerprints classification method described in [5,15] to tackle the problem of Event detection. In [15] the authors approach the problem of text authorship detection by using the crime scene fingerprint analogy to claim that a given text has its authors writing style embedded in it.…”

Section: Fingerprint Event Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Textual Event Detection Using Fuzzy Fingerprints

Marujo

Carvalho

Gershman

et al. 2015

Advances in Intelligent Systems and Computing

Self Cite

View full text Add to dashboard Cite

Abstract. In this paper we present a method to improve the automatic detection of events in short sentences when in the presence of a large number of event classes. Contrary to standard classification techniques such as Support Vector Machines or Random Forest, the proposed Fuzzy Fingerprints method is able to detect all the event classes present in the ACE 2005 Multilingual Corpus, and largely improves the obtained G-Mean value.

show abstract

Section: Introductionmentioning

confidence: 70%

Section: Fingerprint Event Detectionmentioning

confidence: 99%

Textual Event Detection Using Fuzzy Fingerprints

Marujo

Carvalho

Gershman

et al. 2015

Advances in Intelligent Systems and Computing

Self Cite

View full text Add to dashboard Cite

show abstract

“…It uses a topic modelling approach that combines token unigrams and character four-grams with exogenous features, such as the content of Web pages that correspond to embedded URLs. [34] addresses Topic Detection 16 in the context of Twitter using fuzzy fingerprints: every topic (i.e., hashtag) is represented by the set of its top-k most frequent token unigrams, which are used to estimate its similarity with every new, unclassified tweet. A system for "trend stuffing" is proposed in [21]; it employs a binary classification scheme to decide whether a tweet is related to a highly active topic ("trend") so as to facilitate the detection of spam tweets.…”

Section: Related Workmentioning

confidence: 99%

“…Topic Detection is similar to Topic Classification, but differs in that it involves many more classes, which are also so rare that an unlabelled document is likely to belong to none of them[34].…”

mentioning

confidence: 99%

Graph vs. bag representation models for the topic classification of web documents

2015

View full text Add to dashboard Cite

Text classification constitutes a popular task in Web research with various applications that range from spam filtering to sentiment analysis. In this paper, we argue that its performance depends on the quality of Web documents, which varies significantly. For example, the curated content of news articles involves different challenges than the usergenerated content of blog posts and Social Media messages. We experimentally verify our claim, quantifying the main factors that affect the performance of text classification. We also argue that the established bag-of-words representation models are inadequate for handling all document types, as they merely extract frequent, yet distinguishing terms from the textual content of the training set. Thus, they suffer from low robustness in the context of noisy or unseen content, unless they are enriched with contextual, application-specific information. In their place, we propose the use of n-gram graphs, a model that goes beyond the bag-of-words representation, transforming every document into a graph: its nodes correspond to character or word n-grams and the co-occurring ones are connected by weighted edges. Individual document graphs can be combined into class graphs and graph similarities are employed to position and classify documents into the vector space. This approach offers two advantages with respect to bag models: first, classification accuracy increases due to the contextual information that is encapsulated in the edges of the n-gram graphs. Second, World Wide Web it reduces the search space to a limited set of robust, endogenous features that depend on the number of classes, rather than the size of the vocabulary. Our thorough experimental study over three large, real-world corpora confirms the superior performance of n-gram graphs across the main types of Web documents.

show abstract

“…Using a Topic Detection algorithm [1], we obtained an additional 25757 unhastagged tweets about the London Riots. It consists of a Twitter Topic Fuzzy Fingerprint algorithm [14] that provides a weighted rank of keywords for each topic in order to identify a smaller subset of tweets within scope. This method has proven to achieve better results than other well known classifiers in the context of detecting Topics within Twitter, while also being faster in execution.…”

Section: Datasetmentioning

confidence: 99%

Page Rank Versus Katz: Is the Centrality Algorithm Choice Relevant to Measure User Influence in Twitter?

Rosa

Carvalho

Astudillo

et al. 2018

Interactions Between Computational Intelligence and Mathematics

Self Cite

View full text Add to dashboard Cite

Microblogs, such as Twitter, have become an important sociopolitical analysis tool. One of the most important tasks in such analysis is the detection of relevant actors within a given topic through data mining, i.e., identifying who are the most influential participants discussing the topic. Even if there is no gold standard for such task, the adequacy of graph based centrality tools such as PageRank and Katz is well documented. In this paper, we present a case study based on a "London Riots" Twitter database, where we show that Katz is not as adequate for the task of important actors detection since it fails to detect what we refer to as "indirect gloating", the situation where an actor capitalizes on other actors referring to him.

show abstract

Twitter Topic Fuzzy Fingerprints

Cited by 30 publications

References 12 publications

Textual Event Detection Using Fuzzy Fingerprints

Textual Event Detection Using Fuzzy Fingerprints

Graph vs. bag representation models for the topic classification of web documents

Page Rank Versus Katz: Is the Centrality Algorithm Choice Relevant to Measure User Influence in Twitter?

Contact Info

Product

Resources

About