S2S: structural-to-syntactic matching similar documents

Aygün, Ramazan

doi:10.1007/s10115-007-0108-0

Cited by 13 publications

(5 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, α m )). 1 If the j-th token in the entire token space appears in an entity e i , then α j is the TF/IDF 2 weight value of the j-th token. Otherwise, α j = 0.…”

Section: Graph Formationmentioning

confidence: 99%

“…Our problem is associated with the entity resolution (ER) problem that has been known as various names-record linkage [12,15], citation matching [25,32], identity uncertainty [32], merge/purge [19], object matching [1,8,24,39], duplicate detection [41], group linkage [28,29] and so on. Recently, as one of specialized ER problems, [23] introduced the mixed entity resolution problem in which instances of different entities Footnote 4 continued web pages and 13 true clusters in the "Davaid Mulford" name data set.…”

Section: Related Workmentioning

confidence: 99%

“…More specifically, we describe the algorithm details in Algorithm 4.1.The multi-level graph partitioning (MGP) algorithm consists of three steps: scaling-down, partitioning, and scaling-up. During the scaling-down step, the size of the graph is repeatedly decreased; in the clustering step, the smallest graph is partitioned; and during the scaling-up1 Let us assume that a standard stemming process has been done to filter out stop words[21] 2. By definition, a weight is a certain value normalized in terms of importance of a word token in a document.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Scalable clustering methods for the name disambiguation problem

Lee

2011

Knowl Inf Syst

View full text Add to dashboard Cite

When non-unique values are used as the identifier of entities, due to their homonym, confusion can occur. In particular, when (part of) "names" of entities are used as their identifier, the problem is often referred to as a name disambiguation problem, where goal is to sort out the erroneous entities due to name homonyms (e.g., If only last name is used as the identifier, one cannot distinguish "Masao Obama" from "Norio Obama"). In this paper, in particular, we study the scalability issue of the name disambiguation problemwhen (1) a small number of entities with large contents or (2) a large number of entities get un-distinguishable due to homonyms. First, we carefully examine two of the state-of-the-art solutions to the name disambiguation problem and point out their limitations with respect to scalability. Then, we propose two scalable graph partitioning algorithms known as multi-level graph partitioning and multi-level graph partitioning and merging to solve the large-scale name disambiguation problem. Our claim is empirically validated via experimentation-our proposal shows orders of magnitude improvement in terms of performance while maintaining equivalent or reasonable accuracy compared to competing solutions.

show abstract

“…, α m )). 1 If the j-th token in the entire token space appears in an entity e i , then α j is the TF/IDF 2 weight value of the j-th token. Otherwise, α j = 0.…”

Section: Graph Formationmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Scalable clustering methods for the name disambiguation problem

Lee

2011

Knowl Inf Syst

View full text Add to dashboard Cite

show abstract

“…Syntactic and semantic similarity metrics used by SASMINT are listed below: (a) Syntactic similarity: Among different metrics from NLP available for comparing two strings syntactically, we have selected for SASMINT a number of well-known ones that are suitable for different types of strings. Such metrics are also used in the Information Retrieval (IR) domain for determining the document similarities, as addressed in Aygün [6], Wan [63]. Combining the results of syntactic similarity metrics using a weighted summation makes SASMINT applicable for different domains, which typically consist of schema elements in varying forms.…”

Section: Linguistic Matchingmentioning

confidence: 99%

Semi-automated schema integration with SASMINT

Unal

2009

Knowl Inf Syst

View full text Add to dashboard Cite

The emergence of increasing number of collaborating organizations has made clear the need for supporting interoperability infrastructures, enabling sharing and exchange of data among organizations. Schema matching and schema integration are the crucial components of the interoperability infrastructures, and their semi-automation to interrelate or integrate heterogeneous and autonomous databases in collaborative networks is desired. The Semi-Automatic Schema Matching and INTegration (SASMINT) System introduced in this paper identifies and resolves several important syntactic, semantic, and structural conflicts among schemas of relational databases to find their likely matches automatically. Furthermore, after getting the user validation on the matched results, it proposes an integrated schema. SASMINT uses a combination of a variety of metrics and algorithms from the Natural Language Processing and Graph Theory domains for its schema matching. For the schema integration, it utilizes a number of derivation rules defined in the scope of the research work explained in this paper. Furthermore, a derivation language called SASMINT Derivation Markup Language (SDML) is defined for capturing and formulating both the results of matching and the integration that can be further used, for example for federated query processing from independent databases. In summary, the paper focuses on addressing: (1) conflicts among schemas that make automatic schema matching and integration difficult, (2) the main components of the SASMINT approach and system, (3) in-depth exploration of SDML, (4) heuristic rules designed and implemented as part of the schema integration component of the SASMINT system, and (5) experimental evaluation of SASMINT.

show abstract

“…This additional link enables rapid transitions when there is a failure in the pattern matching process, by which the automata can move to another trie branch that has similar prefix without the need for backtracking. Aho-Corasick algorithm has been applied to solve numerous problem such as signature-based anti-virus application [2], set matching in Bioinformatics [4], structural-to-syntactic matching for identical documents [6], searching of text strings on digital forensics [7] and text mining [8].…”

Section: Introductionmentioning

confidence: 99%