Towards Deeper Understanding of the Search Interfaces of the Deep Web

He, Hai; Meng, Weiyi; Lu, Yiyao; Yu, Clement; Wu, Zonghuan

doi:10.1007/s11280-006-0010-9

Cited by 45 publications

(34 citation statements)

References 16 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Considering the subject diversity of deep Web [6], we experimented on interfaces from varied domains, and achieved more than 75% segmentation accuracy in most cases. In biology domain, our approach attained a segmentation accuracy of 87%, achieving 16% improvement over an existing approach, LEX [7]. In automobile domain, our approach attained an accuracy of 79%, achieving 24% improvement over LEX.…”

Section: Figure 1 Segmented Search Interfacementioning

confidence: 83%

“…The work in [22] uses a very similar representation scheme {attribute name, operator, values}. We also share few similarities with LEX [7]; an operand is similar to the domain element, an operator is similar to the constraint element; and a text-based operator, such as between or and (Figure 1), is similar to the element label. Our work differs from the previous ones in that we view the interface design process as a stochastic process for laying out a stream of components with different semantic roles.…”

Section: Figure 2 Assigning Semantic Labels To Componentsmentioning

confidence: 99%

“…The tuples are identified using pre-specified grammar rules. An example of a heuristic-based technique is He et al [7] s LEX. The segment, logical attribute, is identified using a set of heuristics.…”

Section: Interface Segmentationmentioning

confidence: 99%

“…In certain examples, a general attribute-name such as ID, was followed by a specific one such as Gene ID as shown in the top segment of Figure 6, or an attribute-name was repeated as shown in the bottom segment of Figure 6. T is added to the list because an attribute-name is likely to end with a colon; token 9 1 is added because a parenthesized string is not likely to be an attribute-name [7]. …”

Section: Model Specificationmentioning

confidence: 99%

“…In the first part, we measure the segmentation accuracy achieved on using HMMs. We compare the HMM based approach to a previous heuristic-based solution, LEX [7]. We choose LEX because the details of the heuristic-based algorithm are available in the literature and it explicitly studies the segmentation problem.…”

Section: Empirical Studymentioning

confidence: 99%

See 4 more Smart Citations

An empirical study on using hidden markov model for search interface segmentation

Khare

2009

Proceedings of the 18th ACM Conference on Information and Knowledge Management

View full text Add to dashboard Cite

This paper describes a hidden Markov model (HMM) based approach to perform search interface segmentation. Automatic processing of an interface is a must to access the invisible contents of deep Web. This entails automatic segmentation, i.e., the task of grouping related components of an interface together. While it is easy for a human to discern the logical relationships among interface components, machine processing of an interface is difficult. In this paper, we propose an approach to segmentation that leverages the probabilistic nature of the interface design process. The design process involves choosing components based on the underlying database query requirements, and organizing them into suitable patterns. We simulate this process by creating an artificial designer in the form of a 2-layered HMM. The learned HMM acquires the implicit design knowledge required for segmentation. We empirically study the effectiveness of the approach across several representative domains of deep Web. In terms of segmentation accuracy, the HMM-based approach outperforms an existing state-of-the-art approach by at least 10% in most cases. Furthermore, our cross-domain investigation shows that a single HMM trained on data having varied and frequent design patterns can accurately segment interfaces from multiple domains.

show abstract

Section: Figure 1 Segmented Search Interfacementioning

confidence: 83%

Section: Figure 2 Assigning Semantic Labels To Componentsmentioning

confidence: 99%

Section: Interface Segmentationmentioning

confidence: 99%

Section: Model Specificationmentioning

confidence: 99%

Section: Empirical Studymentioning

confidence: 99%

See 3 more Smart Citations

An empirical study on using hidden markov model for search interface segmentation

Khare

2009

Proceedings of the 18th ACM Conference on Information and Knowledge Management

View full text Add to dashboard Cite

show abstract

Multiobjective evolutionary algorithms for context‐based search

Cecchini

Lorenzetti

Maguitman

et al. 2010

J. Am. Soc. Inf. Sci.

View full text Add to dashboard Cite

Formulating high-quality queries is a key aspect of context-based search. However, determining the effectiveness of a query is challenging because multiple objectives, such as high precision and high recall, are usually involved. In this work, we study techniques that can be applied to evolve contextualized queries when the criteria for determining query quality are based on multiple objectives. We report on the results of three different strategies for evolving queries: (a) single-objective, (b) multiobjective with Pareto-based ranking, and (c) multiobjective with aggregative ranking. After a comprehensive evaluation with a large set of topics, we discuss the limitations of the single-objective approach and observe that both the Pareto-based and aggregative strategies are highly effective for evolving topical queries. In particular, our experiments lead us to conclude that the multiobjective techniques are superior to a baseline as well as to well-known and ad hoc query reformulation techniques. IntroductionContext-based search is the process of seeking material based on a topic of interest (Budzik, Hammond, & Birnbaum, 2001;Kraft, Chang, Maghoul, & Kumar, 2006;Maguitman, Leake, & Reichherzer, 2005). Consider, for example, a journalist writing an article about the H1N1 flu pandemic. The journalist has collected a small set of articles related to the topic at hand and would like to retrieve additional material from other sources. This local collection of documents is indexed and tagged as relevant while other documents are added to the index and tagged as irrelevant. The journalist can be assisted by an intelligent system that monitors the journalist's task, generates an initial set of queries, and incrementally refines these queries to better reflect the topic of interest. The initial queries could be generated directly from the journalist's context (e.g., the document that is being edited) or from a short description provided explicitly by the journalist. These initial queries will be incrementally refined based on the small collection of readily available material, which contains documents related to the H1N1 flu pandemic. In subsequent steps, the refined topical queries are used by the system to retrieve relevant material from a larger corpus containing novel material, such as the Web.The availability of powerful search interfaces makes it possible to develop a plethora of applications for information access in context such as the one described earlier. The matching and ranking mechanisms employed by existing search services are commonly fixed by the service provider. The criteria considered for retrieval change from service to service (e.g., content relevance, document popularity, document freshness, meta-information) and are typically obscure to those using the interface. The only access point to relevant material is through the submission of queries. As a consequence, learning to automatically formulate effective topical queries is an important research problem in the area of context-based search. JOURNAL ...

show abstract