Author | Conference | Journal | Organization | Year | DOI
Look for results that meet for the following criteria:
Academic
Publications
Extracting Patterns and Relations from the World Wide Web
Edit
Extracting Patterns and Relations from the World Wide Web   (Citations: 353)
BibTex | RIS | RefWorks Download
The World Wide Web is a vast resource for information. At the same time it is extremely distributed. A particular type of data such as restaurant lists may be scattered across thousands of independent information sources in many different formats. In this paper, we consider the problem of extracting a relation for such a data type from all of these sources automatically. We present a technique which exploits the duality between sets of patterns and relations to grow the target relation starting ...
Conference: International Workshop on the Web and Databases - WebDB , pp. 172-183, 1998
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...In a pattern-based system (e.g., [6, 2, 5, 8]), seed facts like (Germany, FIFA_World_Cup), as an instance of the teamWonTrophy relation between soccer teams and trophies (which was true in 1974 and 1990), can automatically detect textual patterns like “X won the final and became the Y champion” which in turn can discover new facts such as (Spain, FIFA_World_Cup) (which is true for 2010)...
    • ...[6, 2, 12, 16, 5, 7, 36, 8]) is bootstrapped with seed facts for given relations and automatically iterates, in an almost unsupervised manner, between collecting text patterns that contain facts and finding new fact candidates that co-occur with patterns...
    • ...While such feedback loops are well studied for knowledge-harvesting methods that are exclusively pattern-based, such as [6, 2, 5], our approach distinguishes itself from that previous work by including the reasoning phase in each iteration...

    Ndapandula Nakashole, et al. Scalable knowledge harvesting with high precision and high recall

    • ...[13,22,26,40,45], semi-supervised methods, e.g. [1,8], and unsupervised method, e.g...

    Cristina Giannone, et al. Supervised semantic relation mining from linguistically noisy text doc...

    • ...NLP techniques have been used mainly for named-entity tagging of fixed number of classes or for question-answering of specific question types [39, 11, 22, 40, 8]. Wrapper induction tools generate delimiter-based rules derived from training data [9, 14, 24, 7]. The generated wrappers usually heavily rely on the HTML encoding present in the training data...
    • ...dependent on HTML tagging from training corpus) Brin in [9] introduced an algorithm to extract simple relations from the Web that are similar to a small “training set” of pairs (e.g...

    Michael Gubanov, et al. READFAST: Browsing large documents through unified famous objects (UFO...

    • ...These systems typically build on the paradigm of bootstrapping of entity pairs and patterns as proposed by Brin[1]...
    • ...Bootstrapping-based relation extraction [1,3,4,5,6] leverage large amounts of data on the Web efficiently...
    • ...Sergey Brin propose DIPRE system [1] to extract author–book relation form the Web; The Snowball system[3] extracts entity pairs including a predefined relation from a corpus...
    • ...Furthermore, both DIPRE [1] and SatSnowball use a general form to represent extracted patterns...

    Haibo Li, et al. Using Graph Based Method to Improve Bootstrapping Relation Extraction

    • ...On the other hand, patterns can be learned implicitly in an iterative process, as in DIPRE [6] and Snowball [3]...
    • ...While no formal principles exist, the informal insight of Pattern-Relation Duality (or PR Duality) has long been observed since DIPRE [6]...
    • ...Extracting tuples of a given relation from a text corpus has long been studied [6, 3, 9]. However, its dual problem of searching textual patterns only exists implicitly as an intermediate step of tuple extraction...
    • ...For problem setting, we rely on bootstrapping using seed tuples, similar to [6, 3]...
    • ...As input, like many existing pattern-based extraction efforts [6, 3], we assume a small number of seed tuples (e.g., {(Ottawa, Canada), (Beijing, China)}), and our ultimate goal is to find the matching relation (e.g., tuples for capital-city-of)...
    • ...As a concluding remark, the conceptual model PRDualRank not only exemplifies the original PR Duality in [6], but also formally quantifies and thus “rediscovers” it (as first stated in Sect...

    Yuan Fang, et al. Searching patterns for relation extraction over the web: rediscovering...

Order by:

Citations (353)

Share this on | Contribute to Academic