Collective Information Extraction with Context-Specific Consistencies

Abstract

Abstract. Conditional Random Fields (CRFs) have been widely used for information extraction from free texts as well as from semi-structured documents. Interesting entities in semi-structured domains are often consistently structured within a certain context or document. However, their actual compositions vary and are possibly inconsistent among different contexts. We present two collective information extraction approaches based on CRFs for exploiting these context-specific consistencies. The first approach extends linear-chain CRFs by additional factors specified by a classifier, which learns such consistencies during inference. In a second extended approach, we propose a variant of skip-chain CRFs, which enables the model to transfer long-range evidence about the consistency of the entities. The practical relevance of the presented work for real-world information extraction systems is highlighted in an empirical study. Both approaches achieve a considerable error reduction.

Assessing the Impact of Single and Pairwise Slot Constraints in a Factor Graph Model for Template-Based Information Extraction

“…Methodologically, our work is similar to collective information extraction with undirected graphical models as proposed by Bunescu et al [4] or Kluegl et al [9]; however, these approaches are limited to problems of text segmentation, entity tagging and extraction of individual relations.…”

Subgroup discovery

“…It is typically applied, for example, in order to obtain an overview on the relations in the data and for automatic hypotheses generation. Furthermore, also predictive tasks can be tackled, for example, by stacking approaches, or by applying the LeGo framework for combining local patterns into global models.…”

Text mining to detect indications of fraud in annual reports worldwide

“…Also, references from the same source usually have the same ordering of the elements. Anzaroot and McCallum (2013) and Kluegl et al (2012) used conditional random fields (CRF's) instead of HMM's because they offer more flexibility in the design of the input features. Anzaroot and McCallum (2013) applied their model to 1.800 citations from PDF research papers on physics, mathematics, computer science and quantitative biology.…”

“…By automatically segmenting bibliographical references the research literature can be organized more easily to provide insight into the landscape of science and specific research areas. Kluegl et al (2012) make use of context specific information, which is information that documents have in common as a result of the process in which the document is created, such as filling in templates. Adding this type of information improves the results of segmenting bibliographical references.…”

“…However, the texts of CV's and job postings are a little more extensive. Kluegl et al (2012) were able the apply their CRF model to identify the time span and company for which the author of the CV worked. Mooney (1999) developed a model for finding information in newsgroup messages or web pages.…”