Ali Mohammad Zareh Bidoki, Nasser Yazdani, Pedram Ghodsnia
This paper explores the application of recurrent neural networks for the task of robust text classification of a real-world benchmarking corpus. There are many well-established approaches which are used for text classification, but they fail to address the challenge from a more multi-disciplinary viewpoint such as natural language processing and artificial intelligence. The results demonstrate that these recurrent neural networks can be a viable addition to the many techniques used in web intelligence for tasks such as context sensitive email classification and web site indexing.
{"title":"Recurrent Neural Networks for Robust Real-World Text Classification","authors":"Ali Mohammad Zareh Bidoki, Nasser Yazdani, Pedram Ghodsnia","doi":"10.1109/WI.2007.91","DOIUrl":"https://doi.org/10.1109/WI.2007.91","url":null,"abstract":"This paper explores the application of recurrent neural networks for the task of robust text classification of a real-world benchmarking corpus. There are many well-established approaches which are used for text classification, but they fail to address the challenge from a more multi-disciplinary viewpoint such as natural language processing and artificial intelligence. The results demonstrate that these recurrent neural networks can be a viable addition to the many techniques used in web intelligence for tasks such as context sensitive email classification and web site indexing.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"248 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124732210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Web provides easy access to a vast amount of informational content to the average person, who may often be interested in selecting Websites that best match their learning objectives and comprehensibility level. Web content is generally not tagged for easy determination of its instructional appropriateness and comprehensibility level. Our research develops an analytical model, using a group of website features, to automatically determine the comprehensibility level of a Website. These features, selected from a large pool of Website features quantitatively measured, are statistically shown to be significantly correlated to website comprehensibility based on empirical studies. The automatically inferred comprehensibility index may be used to assist the average person, interested in using web content for self-directed learning, to find content suited to their comprehension level and filter out content which may have low potential instructional value.
{"title":"Automatic Website Comprehensibility Evaluation","authors":"Ping Yan, Zhu Zhang, Ray Garcia","doi":"10.1109/WI.2007.27","DOIUrl":"https://doi.org/10.1109/WI.2007.27","url":null,"abstract":"The Web provides easy access to a vast amount of informational content to the average person, who may often be interested in selecting Websites that best match their learning objectives and comprehensibility level. Web content is generally not tagged for easy determination of its instructional appropriateness and comprehensibility level. Our research develops an analytical model, using a group of website features, to automatically determine the comprehensibility level of a Website. These features, selected from a large pool of Website features quantitatively measured, are statistically shown to be significantly correlated to website comprehensibility based on empirical studies. The automatically inferred comprehensibility index may be used to assist the average person, interested in using web content for self-directed learning, to find content suited to their comprehension level and filter out content which may have low potential instructional value.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115102538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Many web pages and resources are primarily relevant to certain geographic locations. For example, in many queries web pages on restaurants, hotels, or movie theaters are mostly relevant to those users who are in geographic proximity to these locations. Moreover, as the number of queries with a local component increases, searching for web pages which are relevant to geographic locations is becoming increasingly important. The performance of geographically-oriented search is greatly affected by how we use geographic information to rank web pages. In this paper, we study the issue of ranking web pages using geographically-sensitive link analysis algorithms. More precisely, we study the question of whether geographic information can improve search performance. We propose several geographically-sensitive link analysis algorithms which exploit the geographic linkage between pages. We empirically analyze the performance of our algorithms.
{"title":"Geographically-Sensitive Link Analysis","authors":"Hyun Chul Lee, Haifeng Liu, Renée J. Miller","doi":"10.1109/WI.2007.134","DOIUrl":"https://doi.org/10.1109/WI.2007.134","url":null,"abstract":"Many web pages and resources are primarily relevant to certain geographic locations. For example, in many queries web pages on restaurants, hotels, or movie theaters are mostly relevant to those users who are in geographic proximity to these locations. Moreover, as the number of queries with a local component increases, searching for web pages which are relevant to geographic locations is becoming increasingly important. The performance of geographically-oriented search is greatly affected by how we use geographic information to rank web pages. In this paper, we study the issue of ranking web pages using geographically-sensitive link analysis algorithms. More precisely, we study the question of whether geographic information can improve search performance. We propose several geographically-sensitive link analysis algorithms which exploit the geographic linkage between pages. We empirically analyze the performance of our algorithms.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116085251","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Due to the proliferation and highly dynamic nature of the Web, an efficient crawling and ranking algorithm for retrieving the most important pages has remained as a challenging issue. Several algorithms like PageRank (Page et al., 1998) and OPIC (Abiteboul et al., 2003) have been proposed. Unfortunately, they have high time complexity. In this paper, an intelligent crawling algorithm based on reinforcement learning, called FICA is proposed that models a real surfing user. The priority for crawling pages is based on a concept which we name as logarithmic distance. FICA is easy to implement and its time complexity is 0(E*logV) where V and E are the number of nodes and edges in the Web graph respectively. Comparison of the FICA with other proposed algorithms shows that FICA outperforms them in discovering highly important pages. Furthermore FICA computes the importance (ranking) of each page during the crawling process. Thus, we can also use FICA as a ranking method for computation of page importance. We have used UK's Web graph for our experiments.
由于Web的扩散和高度动态性，检索最重要页面的有效爬行和排序算法仍然是一个具有挑战性的问题。PageRank (Page et al.， 1998)和OPIC (Abiteboul et al.， 2003)等算法已经被提出。不幸的是，它们具有很高的时间复杂度。本文提出了一种基于强化学习的智能爬行算法(FICA)，对真实的上网用户进行建模。抓取页面的优先级是基于我们称之为对数距离的概念。FICA易于实现，其时间复杂度为0(E*logV)，其中V和E分别为Web图中的节点数和边数。FICA与其他算法的比较表明，FICA在发现高度重要页面方面优于其他算法。此外，FICA在抓取过程中计算每个页面的重要性(排名)。因此，我们也可以使用FICA作为计算页面重要性的排序方法。我们在实验中使用了英国的网络图表。
{"title":"FICA: A Fast Intelligent Crawling Algorithm","authors":"Shady Shehata, F. Karray, Mohamed S. Kamel","doi":"10.1109/WI.2007.132","DOIUrl":"https://doi.org/10.1109/WI.2007.132","url":null,"abstract":"Due to the proliferation and highly dynamic nature of the Web, an efficient crawling and ranking algorithm for retrieving the most important pages has remained as a challenging issue. Several algorithms like PageRank (Page et al., 1998) and OPIC (Abiteboul et al., 2003) have been proposed. Unfortunately, they have high time complexity. In this paper, an intelligent crawling algorithm based on reinforcement learning, called FICA is proposed that models a real surfing user. The priority for crawling pages is based on a concept which we name as logarithmic distance. FICA is easy to implement and its time complexity is 0(E*logV) where V and E are the number of nodes and edges in the Web graph respectively. Comparison of the FICA with other proposed algorithms shows that FICA outperforms them in discovering highly important pages. Furthermore FICA computes the importance (ranking) of each page during the crawling process. Thus, we can also use FICA as a ranking method for computation of page importance. We have used UK's Web graph for our experiments.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"51 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117280906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In online resources, such as news and weblogs, authors often extract articles, embed content, and comment on existing articles related to a popular event. Therefore, it is useful if authors can check whether two or more articles share common parts for further analysis, such as cocitation analysis and search result improvement. If articles do have parts in common, we say the content of such articles is event-relevant. Conventional text classification methods classify a complete document into categories, but they cannot represent the semantics precisely or extract meaningful event-relevant content. To resolve these problems, we propose a near-duplicate detection approach for finding event-relevant content in Web documents. The efficiency of the approach and the proposed duplicate set generation algorithms make it suitable for identifying event-relevant content. The experiment results demonstrate the potential of the proposed approach for use in weblogs.
{"title":"Finding Event-Relevant Content from the Web Using a Near-Duplicate Detection Approach","authors":"Hung-Chi Chang, Jenq-Haur Wang, Chih-Yi Chiu","doi":"10.1109/WI.2007.58","DOIUrl":"https://doi.org/10.1109/WI.2007.58","url":null,"abstract":"In online resources, such as news and weblogs, authors often extract articles, embed content, and comment on existing articles related to a popular event. Therefore, it is useful if authors can check whether two or more articles share common parts for further analysis, such as cocitation analysis and search result improvement. If articles do have parts in common, we say the content of such articles is event-relevant. Conventional text classification methods classify a complete document into categories, but they cannot represent the semantics precisely or extract meaningful event-relevant content. To resolve these problems, we propose a near-duplicate detection approach for finding event-relevant content in Web documents. The efficiency of the approach and the proposed duplicate set generation algorithms make it suitable for identifying event-relevant content. The experiment results demonstrate the potential of the proposed approach for use in weblogs.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123505762","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
It is well accepted that ontology is useful for personalized Web information gathering. However, it is challenging to use semantic relations of "kind-of", "part-of", and "related-to" and synthesize commonsense and expert knowledge in a single computational model. In this paper, a personalized ontology model is proposed attempting to answer this challenge. A two-dimensional (Exhaustivity and Specificity) method is also presented to quantitatively analyze these semantic relations in a single framework. The proposals are successfully evaluated by applying the model to a Web information gathering system. The model is a significant contribution to personalized ontology engineering and concept-based Web information gathering in Web Intelligence.
{"title":"Ontology Mining for Personalized Web Information Gathering","authors":"Xiaohui Tao, Yuefeng Li, N. Zhong, R. Nayak","doi":"10.1109/WI.2007.82","DOIUrl":"https://doi.org/10.1109/WI.2007.82","url":null,"abstract":"It is well accepted that ontology is useful for personalized Web information gathering. However, it is challenging to use semantic relations of \"kind-of\", \"part-of\", and \"related-to\" and synthesize commonsense and expert knowledge in a single computational model. In this paper, a personalized ontology model is proposed attempting to answer this challenge. A two-dimensional (Exhaustivity and Specificity) method is also presented to quantitatively analyze these semantic relations in a single framework. The proposals are successfully evaluated by applying the model to a Web information gathering system. The model is a significant contribution to personalized ontology engineering and concept-based Web information gathering in Web Intelligence.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114763704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper proposes a method for the mapping of ontologies that, in a greater extent than other approaches, discovers and exploits sets of latent features for approximating the intended meaning of ontology elements. This is done by applying the reverse generative process of the Latent Dirichlet Allocation model. Similarity between element pairs is computed by means of the Kullback-Leibler divergence measure. Experimental results show the potential of the method.
{"title":"Mapping Ontologies Elements using Features in a Latent Space","authors":"Vassilis Spiliopoulos, G. Vouros, V. Karkaletsis","doi":"10.1109/WI.2007.74","DOIUrl":"https://doi.org/10.1109/WI.2007.74","url":null,"abstract":"This paper proposes a method for the mapping of ontologies that, in a greater extent than other approaches, discovers and exploits sets of latent features for approximating the intended meaning of ontology elements. This is done by applying the reverse generative process of the Latent Dirichlet Allocation model. Similarity between element pairs is computed by means of the Kullback-Leibler divergence measure. Experimental results show the potential of the method.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128165944","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Today's Web pages are commonly made up of more than merely one cohesive block of information. For instance, news pages from popular media channels such as Financial Times or Washington Post consist of no more than 30%-50% of textual news, next to advertisements, link lists to related articles, disclaimer information, and so forth. However, for many search-oriented applications such as the detection of relevant pages for an in-focus topic, dissecting the actual textual content from surrounding page clutter is an essential task, so as to maintain appropriate levels of document retrieval accuracy. We present a novel approach that extracts real content from news Web pages in an unsupervised fashion. Our method is based on distilling linguistic and structural features from text blocks in HTML pages, having a particle swarm optimizer (PSO) learn feature thresholds for optimal classification performance. Empirical evaluations and benchmarks show that our approach works very well when applied to several hundreds of news pages from popular media in 5 languages.
{"title":"Content Extraction from News Pages Using Particle Swarm Optimization on Linguistic and Structural Features","authors":"Cai-Nicolas Ziegler, Michal Skubacz","doi":"10.1109/WI.2007.38","DOIUrl":"https://doi.org/10.1109/WI.2007.38","url":null,"abstract":"Today's Web pages are commonly made up of more than merely one cohesive block of information. For instance, news pages from popular media channels such as Financial Times or Washington Post consist of no more than 30%-50% of textual news, next to advertisements, link lists to related articles, disclaimer information, and so forth. However, for many search-oriented applications such as the detection of relevant pages for an in-focus topic, dissecting the actual textual content from surrounding page clutter is an essential task, so as to maintain appropriate levels of document retrieval accuracy. We present a novel approach that extracts real content from news Web pages in an unsupervised fashion. Our method is based on distilling linguistic and structural features from text blocks in HTML pages, having a particle swarm optimizer (PSO) learn feature thresholds for optimal classification performance. Empirical evaluations and benchmarks show that our approach works very well when applied to several hundreds of news pages from popular media in 5 languages.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132805794","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this research, we proposed and validated an approach to using information visualization to augment search engines in supporting the analysis of business stakeholder information on the Web. We report in this paper findings from a preliminary evaluation comparing a visualization prototype with a traditional method of stakeholder analysis (Web browsing and searching). We found that the prototype achieved a higher perceived usefulness and perceived analysis effectiveness and was perceived favorably in expediting the subjects' decision making and in helping them understand the analysis results. Overall, the proposed approach was found to augment traditional methods of analyzing business stakeholders. We discuss implications to researchers and practitioners and future directions.
{"title":"Supporting Web Searching of Business Intelligence with Information Visualization","authors":"Wingyan Chung, Ada Leung","doi":"10.1109/WI.2007.149","DOIUrl":"https://doi.org/10.1109/WI.2007.149","url":null,"abstract":"In this research, we proposed and validated an approach to using information visualization to augment search engines in supporting the analysis of business stakeholder information on the Web. We report in this paper findings from a preliminary evaluation comparing a visualization prototype with a traditional method of stakeholder analysis (Web browsing and searching). We found that the prototype achieved a higher perceived usefulness and perceived analysis effectiveness and was perceived favorably in expediting the subjects' decision making and in helping them understand the analysis results. Overall, the proposed approach was found to augment traditional methods of analyzing business stakeholders. We discuss implications to researchers and practitioners and future directions.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126595690","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Olga Goussevskaia, Michael Kuhn, Roger Wattenhofer
The virtual world is comprised of data items related to each other in a variety of contexts. Often such relations can be represented as graphs that evolve over time. Examples include social networks, co-authorship graphs, and the world-wide-web. Attempts to model these graphs have introduced the notions of hierarchies and layers, which correspond to taxonomies of the underlying objects, and reasons for object relations, respectively. In this paper we explore these concepts in the process of mining such naturallygrown networks. Based on two sample graphs, we present some evidence that the current models well fit real world networks and provide concrete applications of these findings. In particular, we show how hierarchies can be used for greedy routing and how separation of layers can be used as a preprocessing step to implement a location estimation application.
{"title":"Layers and Hierarchies in Real Virtual Networks","authors":"Olga Goussevskaia, Michael Kuhn, Roger Wattenhofer","doi":"10.1109/WI.2007.69","DOIUrl":"https://doi.org/10.1109/WI.2007.69","url":null,"abstract":"The virtual world is comprised of data items related to each other in a variety of contexts. Often such relations can be represented as graphs that evolve over time. Examples include social networks, co-authorship graphs, and the world-wide-web. Attempts to model these graphs have introduced the notions of hierarchies and layers, which correspond to taxonomies of the underlying objects, and reasons for object relations, respectively. In this paper we explore these concepts in the process of mining such naturallygrown networks. Based on two sample graphs, we present some evidence that the current models well fit real world networks and provide concrete applications of these findings. In particular, we show how hierarchies can be used for greedy routing and how separation of layers can be used as a preprocessing step to implement a location estimation application.","PeriodicalId":192501,"journal":{"name":"IEEE/WIC/ACM International Conference on Web Intelligence (WI'07)","volume":"92 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-11-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115579279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：