Document re-ranking based on topic-comment structure

Ermakova, Liana; Mothe, Josiane

doi:10.1109/rcis.2016.7549352

Cited by 6 publications

(4 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For these inquiries just human mediation was required. The records got after Google looking against each inquiry are positioned utilizing OkapiBM25 [24]. Numerous and assumes a significant part in Q.E.…”

Section: Experimentation and Results Analysismentioning

confidence: 99%

Cloud Based Multi-Language Indexing Using Cross Lingual Information Retrieval Approaches

al.¹

2021

ITII

View full text Add to dashboard Cite

The exponential growth of data sizes created by digital media (video/audio/images), physicalsimulations, scientific instruments and web authoring joins the new growth of interest in cloud computing. The options for distribution and parallelization of information in clouds make the retrieval and storage processes very complicated, especially when faced with real-time data management. The quantity of Web Users getting access to data over Internet is expanding step by step. An enormous measure of data on Internet is accessible in various languages which could be accessed by anyone whenever. The Information Retrieval (IR) manages finding valuable data from a huge assortment of unorganized, organized and semi-organized information. In the present situation, the variety of data and language boundaries are the difficult challenges for communication and social trade over the world. To tackle such obstructions, CLIR, the cross-language information retrieval frameworks, are these days in solid interest. The Query Expansion (Q.E.) is the way toward adding related and important terms to original inquiry to upgrade its indexing ability to improve the significance of recovered files in CLIR. In this exploration work, Q.E. has been investigated for a Hindi-English and Kannada-English CLIR in that Hindi and Kannada queries are utilized to look through English docs. After the interpretation of query, recovered outcomes are positioned making use of OkapiBM25 to organize the most important doc at the top for expanding the significance of recovered docs using QE. We proposed architecture for Hindi-English and Kannada-English CLIR making use of QE. to improve the importance of recovered reports. In the primary investigation, QE. is performed with and without OkapiBM25 ranking. The outcomes show that the pertinence of recovered archives is higher with OKapiBM25 as contrast with the one without positioning. The work docs plainly demonstrate that the presentation of Hindi-English and Kannada-English CLIR framework can be improved altogether with query development using fitting terms located at suitable place and the recovered Snippets can incredibly fill in as the continuous test collection.

show abstract

Section: Experimentation and Results Analysismentioning

confidence: 99%

Cloud Based Multi-Language Indexing Using Cross Lingual Information Retrieval Approaches

al.¹

2021

ITII

View full text Add to dashboard Cite

show abstract

“…ese parsers are based on deep analysis of linguistic features and are hardly usable when large volumes of texts are involved. Moreover, they view the topiccomment relation as a remark on the statement while we consider a topic as the phrase that the rest of the clause is understood to be about as in [13].…”

Section: Topic-comment Structure For Somentioning

confidence: 99%

“…Since the unmarked word order in English is Subject -Verb -Object (SVO), we can assume that a topic is usually placed before a verb. As in [13], we also assume that if a subordinate clause provides details on an object, it is rather related to a comment part. us, in our method we split a sentence into two parts by a personal verb (not in nitive nor participle) where the rst part is considered to be a topic while the rest is viewed as a comment.…”

Section: Topic-comment Structure For Somentioning

confidence: 99%

“…Although, in literature topic-comment structure has been exploited for document re-ranking [13], classi cation [5], and text summarization [11], to our knowledge, it has never been applied for SO. e contribution of this paper is a completely automatic approach for SO evaluation based on topic-comment structure of a text that requires only shallow parsing and has linear complexity.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Metric for Sentence Ordering Assessment Based on Topic-Comment Structure

Ermakova

Mothe

Firsov

2017

Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

View full text Add to dashboard Cite

Sentence ordering (SO) is a key component of verbal ability. It is also crucial for automatic text generation. While numerous researchers developed various methods to automatically evaluate the informativeness of the produced contents, the evaluation of readability is usually performed manually. In contrast to that, we present a selfsu cient metric for SO assessment based on text topic-comment structure. We show that this metric has high accuracy.

show abstract

A Normalized Framework Based on Multiple Relationships for Document Re-ranking

Zhao

Zhou

2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Document re-ranking based on topic-comment structure

Cited by 6 publications

References 21 publications

Cloud Based Multi-Language Indexing Using Cross Lingual Information Retrieval Approaches

Cloud Based Multi-Language Indexing Using Cross Lingual Information Retrieval Approaches

A Metric for Sentence Ordering Assessment Based on Topic-Comment Structure

A Normalized Framework Based on Multiple Relationships for Document Re-ranking

Contact Info

Product

Resources

About