Figure 9 - uploaded by Payal Gulati
Content may be subject to copyright.
Comparison of Prevailing and Proposed System
Source publication
As more information becomes available on the Web it is moredifficult to provide effective search services for Internet users.Since, it is assumed that users do not always formulate searchqueries using the best terms. So, search engines invoke queryexpansion to increase the quality of user search results. Queryexpansion is useful in reducing query/d...
Context in source publication
Similar publications
In this world of fast computing, automation plays an important role. In image retrieval technique automation is a great quest. Giving an image as a query and retrieving relevant images is a challenging research area. In this paper we are proposing a research of using Multi-Modality Ontology integration for image retrieval concept. The core strategy...
Searching useful information from unstructured medical multimedia data has been a difficult problem in information retrieval. This paper reports an effective semantic medical multimedia retrieval approach which can reflect the users' query intent. Firstly, semantic annotations will be given to the multimedia documents in the medical multimedia data...
Citations
... One effective way to cope with the FAP is the query expansion. Query expansion has proved to improve the effectiveness of retrieval by automatically enhancing terms used in original query with other related terms while maintaining the user intention [10,33]. ...
Annotation-based image retrieval associates textual descriptions to images based on human perception. A user query, composed of keywords of choice and for retrieval, are usually matched lexically with the textual descriptions associated for stored images to extract the best matches. This paradigm will not produce appropriate desired results for complex queries if a semantic approach is not considered. This paper proposes an image retrieval framework which integrates external knowledge sources for obtaining a higher-level inference that can both handle complex queries and increase the number of relevant retrievals. The framework includes a parser where a semantic representation graph is initially generated from both image captions and query. The semantic representation of image captions is stored in the form of Resource Description Framework (RDF) triples, while the user query is translated into a SPARQL language query. For better query understanding, the external knowledge sources (ConceptNet, WordNet), are next fused together with the parser’s output in a significant process named query expansion to infer combined and expanded knowledge about the terms used in the query. Also, the expansion process generates a set of expansion rules to semantically expand the user query to adapt the inferred knowledge. The expanded query is matched against the stored RDF triplets to indicate the best matched image retrievals. Retrievals are eventually ranked using a relation similarity metric to obtain a ranked list of relevant images. Experimental studies carried on two Flickr datasets show that the proposed framework outperforms related work with 40% increase in the number of relevant retrievals at almost full accuracy. The framework achieves additionally an average increase for the accuracy at given k in the range of 50–72% for up to the tenth retrieval.
... Other researchers are trying to exploit some semantic resources for the query expansion. In [9], the author uses WordNet 1 for the expansion. The others of [16] proposed some similarity measures based on WordNet and Wikipedia to extend queries with the similar concepts. ...
In the context-based image retrieval, the textual information surrounding the image plays a central role for ranking returned results. Although this technique outperforms content-based approaches, it may fail when the query keywords does not match the textual content of many documents containing relevant images. In addition, users are usually not experts and provide ambiguous queries that lead to heterogeneous results. To solve these problems, researchers are trying to re-rank primary results using other techniques such as query expansion, concept-based retrieval, etc. In this paper, we propose to use LDA topic model to re-rank results and therefore improve retrieval precision. We apply this model in two levels: global level represented by the whole document containing the image and local level represented by the paragraph containing an image (considered as a specific textual information for the image). Results show a significant improvement over the standard text retrieval approach by re-ranking with the LDA model applied to the local level.
... They performed phrase recognition and define WSD on queries and then selected highly correlated terms of the same sense with query terms. Similarly different researchers have been using the WordNet for information retrieval like from a mini web logs image retrieval [8], log mining techniques for support Web Query [12], Folksonomy tag co-occurrence for Web [13], Compact Concept Ontology (CCO) [22], Abouenour et al [23] have expanded the query of user through WordNet for Arabic language and get a high-precision result from the Retrieval system. Wang and Zhang [11] have expanded the user query through WordNet and then remove the semantic similarity space between the expanded terms and original query. ...
... The similarity between the filter expanded terms and the images along with user original query will be calculated in the form of cosine measurement. cosΘIi=similarity(Q', Ii) (8) Similarity (Q', Ii) =∑ , , ...
-With the tremendous revolutions in the digital technology and ease of use in the
results large volumes of images generated. In such types gigantic image corpus it is hardly
to find the relevant images. Different researchers have defined techniques to retrieve images
with high accuracy but still not cope in the semantic gap and lexical ambiguity. This paper
focused on short query expansion ie expands user query with ConceptNet and WordNet
knowledge bases. It retrieved the images, even if they are labeled with different words. .
This chapter explains the multi-agent system for effective information retrieval using information scent in query log mining. The precision of search results is low due to difficult to infer the information need of the small size search query and therefore information need of the user is not satisfied effectively. Information Scent is used for modeling the information need of user web search session and clustering is performed to identify the similar information need sessions. Hyper Link-Induced Topic Search (HITS) is executed on clusters to generate the Hubs and authorities for web page recommendations to users who search with similar intents. This multi-agent system based on clustered query sessions uses query operations like expansion and recommendation to infer the information need of user search queries and recommends Hubs and authorities for effective web search.
This chapter explains the multi-agent system for effective information retrieval using information scentin query log mining. The precision of search results is low due to difficult to infer the information needof the small size search query and therefore information need of the user is not satisfied effectively.Information Scent is used for modeling the information need of user web search session and clusteringis performed to identify the similar information need sessions. Hyper Link-Induced Topic Search (HITS)is executed on clusters to generate the Hubs and authorities for web page recommendations to userswho search with similar intents. This multi-agent system based on clustered query sessions uses queryoperations like expansion and recommendation to infer the information need of user search queries andrecommends Hubs and authorities for effective web search.
We study several semantic concept-based query expansion and re-ranking scheme and compare different ontology-based expansion methods in image search and retrieval. To improve the query expansion efficiency and accuracy, we employ the CYC knowledge base to generate the expansion candidate concepts, while filter and rank the expansion results by calculating concept similarities using the Semantic Relatedness Metrics. Using our knowledge-based query expansion in image retrieval, the efficiency and accuracy has been improved.
This paper explains the application of ontologies in financial domains to a query expansion process. The final goal is to improve financial information retrieval effectiveness. The system is composed of an ontology and a Lucene index that stores and retrieves natural language concepts. An initial evaluation with a limited number of queries has been performed. Obtained results show that ambiguity remains a problem when expanding a query. The filtering of entities in the expansion process by selecting only companies or references to markets helps in the reduction of ambiguity. © 2013 Sociedad Española Para el Procesamiento del Lenguaje Natural.
The evolution in the digital technology from the last few decades and different multimedia sources like Broadcast news, Movies, Videos, images, etc. have increased the size volume of the digital media daily. Due to this explosive growth in the digital media volume, it’s strongly urged for the system that efficiently and effectively compiles the user demand and retrieving the relevant images. In this paper, the user query will be expanded through an open-source knowledge base WordNet and ConceptNet for retrieves the images on the base of different synonyms and concepts. This technique covers the word mismatch and word sense disambiguation (WSD) problem. Propose method effectively applied on the open benchmark image dataset LabelMe. The experimental results show that of the propose techniques have improved the retrieval performance over the traditional ones and get the accuracy up to 94% for single word and 83% for sentence word queries.
