Extracting significant time varying features from text

Swan, Russell; Allan, James

doi:10.1145/319950.319956

Cited by 81 publications

(56 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Intuitively, burst detection can be achieved by identifying a burst region where the data value exceeds certain threshold. The threshold can be determined based on heuristics [20], different data distribution assumptions [21] or statistical tests [22]. The Cumulative Sum (CUSUM) method [23] is one of the most popular statistical approach for change point detection.…”

Section: Related Workmentioning

confidence: 99%

Tracing Coordination and Cooperation Structures via Semantic Burst Detection

Lin¹,

Margolin²,

Lazer³

2014

EAI Endorsed Transactions on Collaborative Computing

View full text Add to dashboard Cite

Developing technologies that support collaboration requires understanding how knowledge and expertise are shared and distributed among community members. We explore two forms of knowledge distribution structures, coordination and cooperation, that are central to successful collaboration. We propose a novel method for detecting the coordination of strategic communication among members of political communities. Our method identifies a "rapid semantic convergence," a sudden burst in the use linguistic constructions by multiple individuals within a short time, as a signature of coordination. We apply our method to the public statements of U.S. Senators in the 112th U.S. Congress and construct coordination and cooperation networks among these individuals. We then compare aspects of these networks to other known properties of the Senators. Results indicate that the detected networks reflect underlying tendencies in the social relationships among Senators and reveal interesting differences in how the different parties coordinate communication.

show abstract

Section: Related Workmentioning

confidence: 99%

Tracing Coordination and Cooperation Structures via Semantic Burst Detection

Lin¹,

Margolin²,

Lazer³

2014

EAI Endorsed Transactions on Collaborative Computing

View full text Add to dashboard Cite

show abstract

“…Significant solutions range from extracting time-varying features from texts (Swan and Allan, 1999) to constructing timelines for event classification based on word usage statistics (Swan and Jensen, 2000) and personalized newsfeeds based on information novelty (Gabrilovich et al, 2004). In the latter, the inter-and intra-document dynamics of documents is considered to model how information evolves over time from article to article, as well as within individual articles.…”

Section: Previous Researchmentioning

confidence: 99%

Demonstrating conceptual dynamics in an evolving text collection

Daranyi

Wittek

2013

J Am Soc Inf Sci Tec

View full text Add to dashboard Cite

Based on real world user demands, we demonstrate how animated visualisation of evolving text corpora displays the underlying dynamics of semantic content. To interpret the results, one needs a dynamic theory of word meaning. We suggest that conceptual dynamics as the interaction between kinds of intellectual, emotional etc. content, and language, is key for such a theory. We demonstrate our methodology by two-way seriation which is a popular technique to analyse groups of similar instances and their features, as well as the connections between the groups themselves. The two-way seriated data may be visualised as a two-dimensional heat map or as a three-dimensional landscape where colour codes or height correspond to the values in the matrix. In this paper we focus on two-way seriation of sparse data in the Reuters-21568 test collection. To achieve a meaningful visualisation thereof we introduce a compactly supported convolution kernel similar to filter kernels used in image reconstruction and geostatistics. This filter populates the high-dimensional sparse space with values that interpolate nearby elements, and provides insight into the clustering structure. We also extend two-way seriation to deal with online updates of both the row and column spaces, and, combined with the convolution kernel, demonstrate a three-dimensional visualisation of dynamics.

show abstract

“…For example, concept drift in textual data streams can be identified by monitoring word frequencies (Swan and Allan 1999) and the formation of new word clusters (Hsiao and Chang 2008;Spinosa et al 2007). Kifer et al (2004) introduce a more generic approach that uses a two window paradigm to detect changes in feature distribution.…”

Section: Triggered Rebuildmentioning

confidence: 99%

Drift detection using uncertainty distribution divergence

Lindstrom¹,

Namee²,

Delany³

2012

Evolving Systems

View full text Add to dashboard Cite

Data generated from naturally occurring processes tends to be nonstationary. For example, seasonal and gradual changes in climate data and sudden changes in financial data. In machine learning the degradation in classifier performance due to such changes in the data is known as concept drift and there are many approaches to detecting and handling it. Most approaches to detecting concept drift, however, make the assumption that true classes for test examples will be available at no cost shortly after classification and base the detection of concept drift on measures relying on these labels. The high labelling cost in many domains provides a strong motivation to reduce the number of labelled instances required to detect and handle concept drift. Triggered detection approaches that do not require labelled instances to detect concept drift show great promise for achieving this. In this paper we present Confidence Distribution Batch Detection (CDBD), an approach that provides a signal correlated to changes in concept without using labelled data. This signal combined with a trigger and a rebuild policy can maintain classifier accuracy which, in most cases, matches the accuracy achieved using classification error based detection techniques but using only a limited amount of labelled data.

show abstract

Extracting significant time varying features from text

Cited by 81 publications

References 4 publications

Tracing Coordination and Cooperation Structures via Semantic Burst Detection

Tracing Coordination and Cooperation Structures via Semantic Burst Detection

Demonstrating conceptual dynamics in an evolving text collection

Drift detection using uncertainty distribution divergence

Contact Info

Product

Resources

About