A New Deep Neural Network Based Learning to Rank Method for Information Retrieval

The recent technology development fascinates the people towards information and its services. Managing the personal and pubic data is a perennial research topic among researchers. In particular retrieval of information gains more attention as it is important similar to data storing. Clustering based, similarity based, graph based information retrieval systems are evolved to reduce the issues in conventional information retrieval systems. Learning based information retrieval is the present trend and in particular deep neural network is widely adopted due to its retrieval performance. However, the similarity between the information has uncertainties due to its measuring procedures. Considering these issues also to improve the retrieval performance, a hybrid deep fuzzy hashing algorithm is introduced in this research work. Hashing efficiently retrieves the information based on mapping the similar information as correlated binary codes and this underlying information is trained using deep neural network and fuzzy logic to retrieve the necessary information from distributed cloud. Experimental results prove that the proposed model attains better retrieval accuracy and accuracy compared to conventional models such as support vector machine and deep neural network.

Download Full-text

Deep neural network based learning to rank for address standardization

10.1109/rivf51545.2021.9642079 ◽

2021 ◽

Author(s):

Hai-Nam Cao ◽

Viet-Trung Tran

Keyword(s):

Neural Network ◽

Deep Neural Network ◽

Learning To Rank

Download Full-text

Deep Neural Network Regularization for Feature Selection in Learning-to-Rank

IEEE Access ◽

10.1109/access.2019.2902640 ◽

2019 ◽

Vol 7 ◽

pp. 53988-54006 ◽

Cited By ~ 1

Author(s):

Ashwini Rahangdale ◽

Shital Raut

Keyword(s):

Neural Network ◽

Feature Selection ◽

Deep Neural Network ◽

Learning To Rank

Download Full-text

English-Vietnamese Cross-Lingual Paraphrase Identification Using MT-DNN

Engineering, Technology & Applied Science Research ◽

10.48084/etasr.4300 ◽

2021 ◽

Vol 11 (5) ◽

pp. 7598-7604

Author(s):

H. V. T. Chi ◽

D. L. Anh ◽

N. L. Thanh ◽

D. Dinh

Keyword(s):

Neural Network ◽

Information Retrieval ◽

Natural Language ◽

Deep Neural Network ◽

State Of The Art ◽

Language Models ◽

Language Understanding ◽

Cross Language Information Retrieval ◽

Cross Lingual ◽

Cross Language

Paraphrase identification is a crucial task in natural language understanding, especially in cross-language information retrieval. Nowadays, Multi-Task Deep Neural Network (MT-DNN) has become a state-of-the-art method that brings outstanding results in paraphrase identification [1]. In this paper, our proposed method based on MT-DNN [2] to detect similarities between English and Vietnamese sentences, is proposed. We changed the shared layers of the original MT-DNN from original the BERT [3] to other pre-trained multi-language models such as M-BERT [3] or XLM-R [4] so that our model could work on cross-language (in our case, English and Vietnamese) information retrieval. We also added some tasks as improvements to gain better results. As a result, we gained 2.3% and 2.5% increase in evaluated accuracy and F1. The proposed method was also implemented on other language pairs such as English – German and English – French. With those implementations, we got a 1.0%/0.7% improvement for English – German and a 0.7%/0.5% increase for English – French.

Download Full-text

Comprehensive Review on Effectual Information Retrieval of Semantic Drift using Deep Neural Network

Asian Journal of Computer Science and Technology ◽

10.51983/ajcst-2019.8.1.2122 ◽

2019 ◽

Vol 8 (1) ◽

pp. 32-35

Author(s):

A. Uma Maheswari ◽

N. Revathy

Keyword(s):

Neural Network ◽

Information Retrieval ◽

Information Extraction ◽

Deep Neural Network ◽

Research Work ◽

Single Class ◽

High Recall ◽

Comprehensive Review ◽

Substantial Loss ◽

Semantic Lexicons

Semantic drift is a common problem in iterative information extraction. Unsupervised bagging and incorporated distributional similarity is used to reduce the difficulty of semantic drift in iterative bootstrapping algorithms, particularly when extracting large semantic lexicons. In this research work, a method to minimize semantic drift by identifying the (Drifting Points) DPs and removing the effect introduced by the DPs is proposed. Previous methods for identifying drifting errors can be roughly divided into two categories: (1) multi-class based, and (2) single-class based, according to the settings of Information Extraction systems that adopt them. Compared to previous approaches which usually incur substantial loss in recall, DP-based cleaning method can effectively clean a large proportion of semantic drift errors while keeping a high recall.

Download Full-text