The Experts below are selected from a list of 18822 Experts worldwide ranked by ideXlab platform
Zhao Shuining - One of the best experts on this subject based on the ideXlab platform.
-
Document Vector space model construction based on domain ontology
Application Research of Computers, 2013Co-Authors: Hao Wenning, Chen Gang, Jin Dawei, Zhao ShuiningAbstract:As VSM couldn't reveal the same conceptual semantics hidden in different feature words and the latent semantic relations in the Documents and had a low accuracy in Document similarity calculation,this paper presented DOBVSM.This model added the conceptions in domain ontology to the feature word set,and then adjusted the weight of feature words using the relation between conceptions,constructed the Document DOBVSM contains semantic relations.Experimental analysis show that the Documents'similarity computed by DOBVSM was more divergent,closest to the expert evaluation and can perform batter in Document similarity.
-
Document Vector space model construction based on ontology in military training
Journal of Computer Applications, 2012Co-Authors: Zhao ShuiningAbstract:In order to solve the problem of semantic revelance in military training Document,this paper proposed a method to construct Vector space model based on ontology in military training.It mainly introduced to construct index based on ontology in military training and construct Vector space model in base of index constructed before,in which the process of constructing Vector space model mainly included feature extraction,weight calculation and dimensionality reduction.Experimental results show that the text representation of Vector space model based on ontology in military training can solve the problem of semantic revelance.
Chew Lim Tan - One of the best experts on this subject based on the ideXlab platform.
-
Retrieval of machine-printed Latin Documents through Word Shape Coding
Pattern Recognition, 2008Co-Authors: Chew Lim TanAbstract:This paper reports a Document retrieval technique that retrieves machine-printed Latin-based Document images through word shape coding. Adopting the idea of image annotation, a word shape coding scheme is proposed, which converts each word image into a word shape code by using a few shape features. The text contents of imaged Documents are thus captured by a Document Vector constructed with the converted word shape code and word frequency information. Similarities between different Document images are then gauged based on the constructed Document Vectors. We divide the retrieval process into two stages. Based on the observation that Documents of the same language share a large number of high-frequency language-specific stop words, the first stage retrieves Documents with the same underlying language as that of the query Document. The second stage then re-ranks the Documents retrieved in the first stage based on the topic similarity. Experiments show that Document images of different languages and topics can be retrieved properly by using the proposed word shape coding scheme.
-
ICDAR - Automatic Detection of Document Script and Orientation
Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), 2007Co-Authors: Chew Lim TanAbstract:This paper presents an identification technique that automatically detects the underlying script and orientation of scanned Document images. In the proposed technique, Document script and orientation are identified by using the stroke density and distribution, which convert each Document image into a Document Vector. For each script at each orientation, a number of reference Document Vectors are first constructed. Script and orientation of the query Document are then determined according to the similarity between the query Document Vector and multiple pre- constructed reference Document Vectors by using the K-nearest neighbor algorithm. Experiments show that the proposed technique is tolerant to the Document skew and able to detect orientations of Documents of different scripts.
-
ICDAR - Identification of Latin-Based Languages through Character Stroke Categorization
Ninth International Conference on Document Analysis and Recognition (ICDAR 2007), 2007Co-Authors: Chew Lim TanAbstract:This paper presents a language identification technique that detects Latin-based languages of imaged Documents without OCR. The proposed technique detects languages through the word shape coding, which converts each word image into a word shape code and accordingly transforms each Document image into an electronic Document Vector. For each Latin-based language under study, a language template is first constructed through a corpus-based learning process. The underlying language of the query Document is then determined based on the similarity between the query Document Vector and multiple constructed language templates. Compared with the reported methods, the proposed language identification technique is fast, accurate, and tolerant to text segmentation error caused by noise and various types of Document degradation. Experimental results show some promising results.
-
AAAI - Script and language identification in degraded and distorted Document images
2006Co-Authors: Chew Lim TanAbstract:This paper reports a statistical identification technique that differentiates scripts and languages in degraded and distorted Document images. We identify scripts and languages through Document Vectorization, which transforms each Document image into an electronic Document Vector that characterizes the shape and frequency of the contained character and word images. We first identify scripts based on the density and distribution of vertical runs between character strokes and a vertical scan line. Latin-based languages are then differentiated using a set of word shape codes constructed using horizontal word runs and character extremum points. Experimental results show that our method is tolerant to noise, Document degradation, and slight Document skew and attains an average identification rate over 95%.
-
ACM Multimedia - Automatic Document orientation detection and categorization through Document Vectorization
Proceedings of the 14th annual ACM international conference on Multimedia - MULTIMEDIA '06, 2006Co-Authors: Chew Lim TanAbstract:This paper presents an automatic orientation detection and categorization technique that is capable of detecting the orientation of multilingual Documents with arbitrary skew and categorizing Document images according to the underlying languages. We carry out orientation detection and categorization through Document Vectorization, which encodes Document orientation and language information and converts each Document image into an electronic Document Vector through the exploitation of the density and distribution of vertical component runs. For each language of interest, a pair of Vector templates is first constructed through a training process. Orientation and category of the query image are then determined based on distances between the query Document Vector and the constructed Vector templates. Experiments over 492 testing Document images show that the average orientation detection and categorization rates reach up to 97.56% and 99.59%, respectively.
Lalit Kumar Awasthi - One of the best experts on this subject based on the ideXlab platform.
-
enhancing web service clustering using length feature weight method for service description Document Vector space representation
Expert Systems With Applications, 2020Co-Authors: Neha Agarwal, Geeta Sikka, Lalit Kumar AwasthiAbstract:Abstract Due to the rapid growth of web services in repositories, discovering the requisite web service is becoming increasingly cumbersome task. It has raised the demand for efficient web service clustering algorithms. In service repositories, when related web services are stored in a clustered way, it enhances the web service discovery process by reducing search space and time. Many eminent researchers have worked in this field and used the Term Frequency – Inverse Document Frequency (TF-IDF) method for representing web services in Vector space. In general, there are various limitations of the TF-IDF approach i.e. (1) Not efficient for large Documents (2) Position of term and its co-occurrences does not matter (3) Unable to analyze how terms are dispersed in different Documents. In the web service scenario, services are represented in short text form. TF-IDF does not work well in web service representation because of the reason that it is unable to effectively find the importance of a term concerning its occurrence in other Documents. If we compare two service Documents i.e. ‘s1’ and ‘s2’ first having a large and second having small number of terms respectively then TF-IDF does not demonstrate the importance of terms in ‘s1’ as smaller to ‘s2’. Therefore, it is not possible to assign effective weights to the terms. In the lack of effective Vector space representation, the performance of the clustering algorithm also degrades. In this paper, we propose a new approach i.e. LFW+K which is based on Length Feature Weight (LFW) for the Vectorized representation of service followed by K-Means clustering. The proposed approach helps to find the informative term from web service and assigns the term weight accordingly by considering parameters like the dimension of the web service Document, maximum frequency of a term in the Document and occurrences of a term in other Documents. LFW+K is applied on the datasets of real-world web services and the performance is measured using standard measurement criteria (i.e. precision, recall, F1-score, and accuracy). Results of the proposed approach are compared with K-Means clustering on TF-IDF representation method i.e. TF-IDF+K. Results show that the proposed method outperforms the clustering done by using TF-IDF method for Vector space representation of web services.
Kosuke Takano - One of the best experts on this subject based on the ideXlab platform.
-
A framework for a feedback process to analyze and personalize a Document Vector space in a feature extraction model
Information Technology & Management, 2009Co-Authors: Kosuke Takano, Xing Chen, K. MasudaAbstract:In this paper, we present a framework for a feedback process to implement a highly accurate Document retrieval system. In the system, a Document Vector space is created dynamically to implement retrieval processing. The retrieval accuracy of the system depends on the Vector space. When the Vector space is created based on a specific purpose and interest of a user, highly accurate retrieval results can be obtained. In this paper, we present a method for analyzing and personalizing the Vector space according to the purposes and interests of users. In order to optimize the Document Vector space, we defined and implemented functions for the operations of adding, deleting and weighting the terms that were used to create the Vector space. By exploiting effectively and dynamically the classified-Document information related to the queries, our methods allow users to retrieve relevant Documents for their interests and purposes. Even if the search results of the initial retrieval space are not appropriate, by applying the proposed feedback operations, our proposed method effectively improves the search results. We also implemented an experimental search system for semantic Document retrieval. Several experimental results including comparisons of our method with the traditional relevance feedback method is presented to clarify how retrieval accuracy was improved by the feedback process and how accurately Documents that satisfied the purpose and interests of users were extracted.
-
An adaptive search system using heterogeneous Document Vector spaces
2009 IEEE Pacific Rim Conference on Communications Computers and Signal Processing, 2009Co-Authors: Kosuke Takano, Xing Chen, Shuichi Kurabayashi, Yasushi KiyokiAbstract:Conventional database selection algorithms are very helpful in improving search results and reducing network overhead and computation time by cutting off databases that are irrelevant to the queries. However, they have not been designed to adapt to the changing interest and intentions of users in the Document retrieval process. In this paper, we propose an adaptive search system using heterogeneous Document Vector spaces. Our system provides a dynamic construction function of search engine components based on a Vector space model. In order to reflect the user's current working contexts in the components of the search engine, our system selects adaptively pre-defined sets of feature terms with different search domains depending on the user's currently-viewed Documents. By exploiting such adaptive and heterogeneous Document Vector spaces with specific domains, our system allows users to retrieve proper Documents that match their current working contexts. Also, our system implements a conventional database selection method to reduce the computation time of the similarity calculations in each Document Vector space. In this study, we confirm that our adaptive search system can improve the precision rates of search results as well as the scalability of computation times, by several experiments using the experimental retrieval system implemented in the desktop environment.
-
SAC - A light-weight feedback method for reconstructing a Document Vector space on a feature extraction model
Proceedings of the 2008 ACM symposium on Applied computing - SAC '08, 2008Co-Authors: Kosuke Takano, Xing ChenAbstract:In this paper, we propose a Document retrieval system with a light-weight feedback method for reconstructing a Document Vector space, which is developed on a Feature Extraction Model (FEM). FEM makes it possible to realize a light-weight creation of Vector spaces by feature terms extracted from the pre-prepared Documents and we can apply the feedback method dynamically to reconstruct the Vector spaces based on intensions of users. Retrieval results can be improved through the proposed feedback process because the distributions of Documents on the reconstructed Vector space are arranged properly according to purposes and interests of users.
-
A Feedback Method of Improving the Performance of a Document Vector Space Developed on a Feature Extraction Model
2007 IEEE Pacific Rim Conference on Communications Computers and Signal Processing, 2007Co-Authors: Kosuke Takano, Chen Xing, K. MasudaAbstract:In this paper, we propose a basic algorithm for an automatic feedback method to reconstruct the Document Vector space that is developed based on a method referred to as feature extraction model (FEM). In the feedback method, we add and remove some terms that are used to construct the Vector space so that retrieval results are improved, because the distributions of Document Vectors are arranged properly according to purposes of searchers. The efficiency and the feasibility of the proposed method are confirmed by several experiments.
K. Masuda - One of the best experts on this subject based on the ideXlab platform.
-
A framework for a feedback process to analyze and personalize a Document Vector space in a feature extraction model
Information Technology & Management, 2009Co-Authors: Kosuke Takano, Xing Chen, K. MasudaAbstract:In this paper, we present a framework for a feedback process to implement a highly accurate Document retrieval system. In the system, a Document Vector space is created dynamically to implement retrieval processing. The retrieval accuracy of the system depends on the Vector space. When the Vector space is created based on a specific purpose and interest of a user, highly accurate retrieval results can be obtained. In this paper, we present a method for analyzing and personalizing the Vector space according to the purposes and interests of users. In order to optimize the Document Vector space, we defined and implemented functions for the operations of adding, deleting and weighting the terms that were used to create the Vector space. By exploiting effectively and dynamically the classified-Document information related to the queries, our methods allow users to retrieve relevant Documents for their interests and purposes. Even if the search results of the initial retrieval space are not appropriate, by applying the proposed feedback operations, our proposed method effectively improves the search results. We also implemented an experimental search system for semantic Document retrieval. Several experimental results including comparisons of our method with the traditional relevance feedback method is presented to clarify how retrieval accuracy was improved by the feedback process and how accurately Documents that satisfied the purpose and interests of users were extracted.
-
A Feedback Method of Improving the Performance of a Document Vector Space Developed on a Feature Extraction Model
2007 IEEE Pacific Rim Conference on Communications Computers and Signal Processing, 2007Co-Authors: Kosuke Takano, Chen Xing, K. MasudaAbstract:In this paper, we propose a basic algorithm for an automatic feedback method to reconstruct the Document Vector space that is developed based on a method referred to as feature extraction model (FEM). In the feedback method, we add and remove some terms that are used to construct the Vector space so that retrieval results are improved, because the distributions of Document Vectors are arranged properly according to purposes of searchers. The efficiency and the feasibility of the proposed method are confirmed by several experiments.