| H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In NIPS, 2002. |
....of such a strategy for obtaining weakly labeled negative examples without user supervision. 2. BACKGROUND The problem of identifying database records that are syntactically different yet describe the same physical entity has been referred to as duplicate detection [13] identity uncertainty [14], object identification [17] and deduplication [15] Record linkage is a variation of the problem that arises when records that describe the same entity are matched across multiple databases [6, 18] Duplicate detection systems go through the process of identifying matching pairs via the ....
.... to evaluate individual approaches, such as: Maximum F measure, which is the harmonic mean between pairwise precision and recall [4, 5, 15] Pairwise precision for the optimal number of pairs [17] Percentage of the correct equivalence classes for which an error exists in the grouping [11, 14]; Proportions of true matching pairs at fixed error levels [19] Although all of these quantities characterize accuracy of duplicate detection systems, they sidestep the problem of selecting the threshold Tsim that separates duplicates from non duplicates. These single value accuracy ....
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In Advances in Neural Information Processing Systems 15. MIT Press. 2003.
....attribute of one mention might combine with the job title attribute of another coreferent mention, and this combination, in turn, might help coresolve a third mention. This has been recognized, and there are some examples of techniques that use heuristic feature merging [Morton, 1997] Recently Pasula et al. 2003] have proposed a formal, relational approach to the problem of identity uncertainty using a type of Bayesian network called a Relational Probabilistic Model [Friedman et al. 1999 ] A great strength of this model is that it explicitly captures the dependence among multiple coreference ....
....problem the number of entities (and thus number of attribute nodes, and the domain of the entity assignment nodes) is unknown. Inference in these models must determine for us the highest probability number of entities. In related work on a generative probabilistic model of identity uncertainty, Pasula et al. 2003] , solve this problem by alternating rounds of Metropolis Hastings sampling on a given model structure with rounds of Metropolis Hastings to explore the space of new graph structures. Our desire to avoid the complexity and lack of scalability of this approach motivates our Model 2. 2.2 Model 2: ....
[Article contains additional citation context not shown here]
Hanna Pasula, Bhaskara Marthi, Brian Milch, Stuart Russell, and Ilya Shpitser. Identity uncertainty and citation matching. In Advances in Neural Information Processing (NIPS), 2003.
....attribute of one mention might combine with the job title attribute of another coreferent mention, and this combination, in turn, might help coresolve a third mention. This has been recognized, and there are some examples of techniques that use heuristic feature merging [Morton, 1997] Recently Pasula et al. 2003] have proposed a formal, relational approach to the problem of identity uncertainty using a type of Bayesian network called a Relational Probabilistic Model [Friedman et al. 1999] A great strength of this model is that it explicitly captures the dependence among multiple coreference decisions. ....
....problem the number of entities (and thus number of attribute nodes, and the domain of the entity assignment nodes) is unknown. Inference in these models must determine for us the highest probability number of entities. In related work on a generative probabilistic model of identity uncertainty, Pasula et al. 2003] , solve this problem by alternating rounds of Metropolis Hastings sampling on a given model structure with rounds of Metropolis Hastings to explore the space of new graph structures. Our desire to avoid the complexity and lack of scalability of this approach motivates our Model 2. 2.2 Model 2: ....
[Article contains additional citation context not shown here]
Hanna Pasula, Bhaskara Marthi, Brian Milch, Stuart Russell, and Ilya Shpitser. Identity uncertainty and citation matching. In Advances in Neural Information Processing (NIPS), 2003.
....papers. This service has had significant impact on the the practice of computer science research. However, the variety of fields and relations it extracts is small, and the limited accuracy of its existing relations constrains the ability to perform more sophisticated data mining. For example, Pasula et al. 2002] note that CiteSeer contains records of over 30 separate AI textbooks written by Russel and Norvig, when actually there is only one. Unfortunately, the complex data mining of rich unstructured text is not feasible with current methods: extraction is often inaccurate, co reference resolution is ....
....have just pointed out, they rely on each other in highly intertwined ways. They cannot be deeply solved separately. This is particularly true of cross document coreference, an extremely important problem that has received little attention. Early work on relational coreference resolution includes Pasula et al. 2002] and McCallum and Wellner [2003] the later is briefly described in section 3.4. 2.3 Fragile data mining One might hope that data mining techniques could compensate for the errors introduced by inaccurate extraction and poor coreference resolution. Research in data mining has a long history of ....
[Article contains additional citation context not shown here]
Hanna Pasula, Bhaskara Marthi, Brian Milch, Stuart Russell, and Ilya Shpitser. Identity uncertainty and citation matching. In Advances in Neural Information Processing (NIPS), 2002.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In NIPS 15. MIT Press, Cambridge, MA, 2003.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In NIPS 15. MIT Press, Cambridge, MA, 2003.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In NIPS, 2002.
No context found.
Pasula, H.; Marthi, B.; Milch, B.; Russell, S.; and Shpitser, I. 2002. Identity uncertainty and citation matching. In Advances in Neural Processing Systems 15. Vancouver, British Columbia: MIT Press.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In Advances in Neural Information Processing (NIPS), 2003.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In NIPS, 2002.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In Advances in Neural Information Processing (NIPS), 2002.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In Advances in Neural Information Processing (NIPS), 2003.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching, 2002.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In Advances in Neural Processing Systems 15, Vancouver, British Columbia, 2002. MIT Press.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In Proc. Advances in Neural Information Processing. MIT Press, 2003.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In Advances in Neural Info. Proc. Systems, 2002.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity uncertainty and citation matching. In NIPS, 2002.
No context found.
H. Pasula, B. Marthi, B. Milch, S. Russell, and I. Shpitser. Identity Uncertainty and Citation Matching. In S. Becker, S. Thrun, and K. Obermayer, editors, Advances in Neural Information Processing Systems 15. MIT Press, 2003.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC