On the Difficulty of Clustering Microblog Texts for Online Reputation Management
Meeting of the Association for Computational LinguisticsPublished 24 June 2011
Fernando Pérez-Téllez, David Pinto, John Cardiff, Paolo Rosso
Citations18
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The aim of this work is to present and compare two different approaches to identify tweets which refer to a company distinguishing them from those which do not, and obtained results are promising while at the same time highlighting the difficulty of this task.
Abstract
In recent years microblogs have taken on an important role in the marketing sphere, in which they have been used for sharing opinions and/or experiences about a product or service. Companies and researchers have become interested in analysing the content generated over the most popular of these, the Twitter platform, to harvest information critical for their online reputation management (ORM). Critical to this task is the efficient and accurate identification of tweets which refer to a company distinguishing them from those which do not. The aim of this work is to present and compare two different approaches to achieve this. The obtained results are promising while at the same time highlighting the difficulty of this task.
Keywords
Computer SciencePhysics and Astronomy
Munich Personal RePEc Archive (Ludwig Maximilian University of Munich)Some methods for classification and analysis of multivariate observations
22,787 Citations1967James B. MacQueen
Lecture notes in computer scienceAn Adapted Lesk Algorithm for Word Sense Disambiguation Using WordNet
850 Citations2002Satanjeev Banerjee, Ted Pedersen
This paper presents an adaptation of Lesk's dictionary-based word sense disambiguation algorithm that uses the lexical database WordNet as the source of glosses for this approach, and attains an overall accuracy of 32%.
Explorations in Automatic Thesaurus Discovery
757 Citations1994Gregory Grefenstette
The aim of this monograph is to provide a catalog of words and phrases used in ThesaurusGeneration, as well as some examples of other writers' work, which have been used in similar contexts.
Short text classification in twitter to improve information filtering
727 Citations2010Bharath Sriram, Dave Fuhry +3 more
A small set of domain-specific features extracted from the author's profile and text is proposed to use to classify short text messages to a predefined set of generic classes such as News, Events, Opinions, Deals, and Private Messages.
Concept based query expansion
717 Citations1993Yonggang Qiu, Hans-Peter Frei
This paper presents a probabilistic query expansion model based on a similarity thesaurus which was constructed automatically and results in a notable improvement in the retrieval effectiveness when measured using both recall-precision and usefulness.
TwitterStand
669 Citations2009Jagan Sankaranarayanan, Hanan Samet +3 more
This work investigates the use of Twitter to build a news processing system, called TwitterStand, from Twitter tweets, to capture tweets that correspond to late breaking news, analogous to a distributed news wire service.
WePS3 Evaluation Campaign: Overview of the On-line Reputation Management Task.
46 Citations2010Enrique Amigó, Javier Artiles +4 more
This paper summarizes the definition, resources, evaluation methodology and metrics, participation and comparative results for the second task of the WEPS-3 evaluation campaign, the so-called OnlineReputation Management task.
The Computer JournalA Self-enriching Methodology for Clustering Narrow Domain Short Texts
38 Citations2010David Pinto, Paolo Rosso +1 more
A self-term expansion methodology for improving the performance of clustering methods when dealing with corpora of this kind is introduced, which allows raw textual data to be enriched by adding co-related terms from an automatically constructed lexical knowledge resource obtained from the same target data set (and not from an external resource).
It was easy, when apples and blackberries were only fruits
30 Citations2010Surender Reddy Yerva, Zoltán Miklós +1 more
The task of classifying Twitter messages, whether they are related to a given company: for example, a set of twitter messages containing a keyword apple is classified, whether a message is related to the company Apple Inc.
ITC-UT: Tweet categorization by query categorization for on-line reputation management
26 Citations2010Minoru Yoshida, Shin Matsushima +3 more
This paper describes the system to categorize each query into 3 or 4 classes according to how much the tweets retrieved by the query contain the “true” entity names that refer to the target entity, and then categorizes each tweet by the rules defined for each class of queries.
Bootstrapping Websites for Classification of Organization Names on Twitter.
15 Citations2010Paul Kalmar
This work aims to reduce noise by collecting only the relevant tweets about a company given the company's website and set of Twitter data, and chooses the co-occurring words in each tweet as the most simple data to acquire.
The University of Amsterdam at WePS3.
15 Citations2010Manos Tsagkias, Krisztian Balog
The focus was on building a general organization classifier that predicts, for each tweet, whether it is about a company, and extracting features that are organization-independent and build on the characteristics of Twitter, such as noisy text, abbreviations and Twitter-specific language.
Lecture notes in computer scienceImproving the Clustering of Blogosphere with a Self-term Enriching Technique
5 Citations2009Fernando Pérez-Téllez, David Pinto +2 more
This paper presents a methodology to improve the performance of clustering techniques on blogs, which does not rely on external resources, and shows that this technique can produce significant improvements in the quality of clusters produced.