Bee-Chung Chen, Lei Chen, R. Ramakrishnan, D. Musicant
In this paper, we introduce a new class of data mining problems called learning from aggregate views. In contrast to the traditional problem of learning from a single table of training examples, the new goal is to learn from multiple aggregate views of the underlying data, without access to the un-aggregated data. We motivate this new problem, present a general problem framework, develop learning methods for RFA (Restriction-Free Aggregate) views defined using COUNT, SUM, AVG and STDEV, and offer theoretical and experimental results that characterize the proposed methods.
{"title":"Learning from Aggregate Views","authors":"Bee-Chung Chen, Lei Chen, R. Ramakrishnan, D. Musicant","doi":"10.1109/ICDE.2006.86","DOIUrl":"https://doi.org/10.1109/ICDE.2006.86","url":null,"abstract":"In this paper, we introduce a new class of data mining problems called learning from aggregate views. In contrast to the traditional problem of learning from a single table of training examples, the new goal is to learn from multiple aggregate views of the underlying data, without access to the un-aggregated data. We motivate this new problem, present a general problem framework, develop learning methods for RFA (Restriction-Free Aggregate) views defined using COUNT, SUM, AVG and STDEV, and offer theoretical and experimental results that characterize the proposed methods.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"492 1","pages":"3-3"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76724219","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 multimedia retrieval, a query is typically interactively refined towards the ‘optimal’ answers by exploiting user feedback. However, in existing work, in each iteration, the refined query is re-evaluated. This is not only inefficient but fails to exploit the answers that may be common between iterations. In this paper, we introduce a new approach called SaveRF (Save random accesses in Relevance Feedback) for iterative relevance feedback search. SaveRF predicts the potential candidates for the next iteration and maintains this small set for efficient sequential scan. By doing so, repeated candidate accesses can be saved, hence reducing the number of random accesses. In addition, efficient scan on the overlap before the search starts also tightens the search space with smaller pruning radius. We implemented SaveRF and our experimental study on real life data sets show that it can reduce the I/O cost significantly.
在多媒体检索中，通过利用用户反馈，查询通常被交互式地细化为“最佳”答案。然而，在现有的工作中，在每次迭代中，精炼的查询都会被重新评估。这不仅效率低下，而且无法利用迭代之间可能常见的答案。本文提出了一种新的相关反馈迭代搜索方法SaveRF (Save random access In Relevance Feedback)。SaveRF预测下一次迭代的潜在候选项，并维护这个小集合以进行有效的顺序扫描。这样做可以节省重复的候选访问，从而减少随机访问的数量。此外，在搜索开始前对重叠部分进行高效扫描，以更小的剪枝半径收紧了搜索空间。我们实现了SaveRF，我们对现实生活数据集的实验研究表明，它可以显着降低I/O成本。
{"title":"SaveRF: Towards Efficient Relevance Feedback Search","authors":"Heng Tao Shen, B. Ooi, K. Tan","doi":"10.1109/ICDE.2006.132","DOIUrl":"https://doi.org/10.1109/ICDE.2006.132","url":null,"abstract":"In multimedia retrieval, a query is typically interactively refined towards the ‘optimal’ answers by exploiting user feedback. However, in existing work, in each iteration, the refined query is re-evaluated. This is not only inefficient but fails to exploit the answers that may be common between iterations. In this paper, we introduce a new approach called SaveRF (Save random accesses in Relevance Feedback) for iterative relevance feedback search. SaveRF predicts the potential candidates for the next iteration and maintains this small set for efficient sequential scan. By doing so, repeated candidate accesses can be saved, hence reducing the number of random accesses. In addition, efficient scan on the overlap before the search starts also tightens the search space with smaller pruning radius. We implemented SaveRF and our experimental study on real life data sets show that it can reduce the I/O cost significantly.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"151 1","pages":"110-110"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76739217","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 of the data sources used in stream query processing are known to exhibit bursty behavior. We focus here on passive network monitoring, an application in which the data rates typically exhibit a large peak-to-average ratio. Provisioning a stream query processor to handle peak rates in such a setting can be prohibitively expensive. In this paper, we propose to solve this problem by provisioning the query processor for typical data rates instead of much higher peak data rates. To enable this strategy, we present mechanisms and policies for managing the tradeoffs between the latency and accuracy of query results when bursts exceed the steady-state capacity of the query processor. We describe the current status of our implementation and present experimental results on a testbed network monitoring application to demonstrate the utility of our approach
{"title":"Declarative Network Monitoring with an Underprovisioned Query Processor","authors":"Frederick Reiss, J. Hellerstein","doi":"10.1109/ICDE.2006.46","DOIUrl":"https://doi.org/10.1109/ICDE.2006.46","url":null,"abstract":"Many of the data sources used in stream query processing are known to exhibit bursty behavior. We focus here on passive network monitoring, an application in which the data rates typically exhibit a large peak-to-average ratio. Provisioning a stream query processor to handle peak rates in such a setting can be prohibitively expensive. In this paper, we propose to solve this problem by provisioning the query processor for typical data rates instead of much higher peak data rates. To enable this strategy, we present mechanisms and policies for managing the tradeoffs between the latency and accuracy of query results when bursts exceed the steady-state capacity of the query processor. We describe the current status of our implementation and present experimental results on a testbed network monitoring application to demonstrate the utility of our approach","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"744 1","pages":"56-56"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76879152","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}
Currently, clinical information is stored in all kinds of proprietary formats through a multitude of medical information systems available on the market. This results in a severe interoperability problem in sharing electronic healthcare records. To address this problem, an industry initiative, called "Integrating Healthcare Enterprise (IHE)" has specified the "Cross Enterprise Document Sharing (XDS)" Profile to store healthcare documents in an ebXML registry/ repository to facilitate their sharing. Through a separate effort, IHE has also defined interdepartmental Workflow Profiles to identify the transactions required to integrate information flow among several information systems. Although the clinical documents stored in XDS registries are obtained as a result of executing these workflows, IHE has not yet specified collaborative healthcare processes for the XDS. Hence, there is no way to track the workflows in XDS and the clinical documents produced through the workflows are manually inserted into the registry/ repository. Given that IHE XDS is using the ebXML architecture, the most natural way to integrate IHE Workflow Profiles to IHE XDS is using ebXML Business Processes (ebBP). In this paper, we describe the implementation of an enhanced IHE architecture demonstrating how ebXML Business Processes, IHE Workflow Profiles and the IHE XDS architecture can all be integrated to provide collaborative business process support in the healthcare domain.
{"title":"Collaborative Business Process Support in IHE XDS through ebXML Business Processes","authors":"A. Dogac, V. Bicer, Alper Okcan","doi":"10.1109/ICDE.2006.39","DOIUrl":"https://doi.org/10.1109/ICDE.2006.39","url":null,"abstract":"Currently, clinical information is stored in all kinds of proprietary formats through a multitude of medical information systems available on the market. This results in a severe interoperability problem in sharing electronic healthcare records. To address this problem, an industry initiative, called \"Integrating Healthcare Enterprise (IHE)\" has specified the \"Cross Enterprise Document Sharing (XDS)\" Profile to store healthcare documents in an ebXML registry/ repository to facilitate their sharing. Through a separate effort, IHE has also defined interdepartmental Workflow Profiles to identify the transactions required to integrate information flow among several information systems. Although the clinical documents stored in XDS registries are obtained as a result of executing these workflows, IHE has not yet specified collaborative healthcare processes for the XDS. Hence, there is no way to track the workflows in XDS and the clinical documents produced through the workflows are manually inserted into the registry/ repository. Given that IHE XDS is using the ebXML architecture, the most natural way to integrate IHE Workflow Profiles to IHE XDS is using ebXML Business Processes (ebBP). In this paper, we describe the implementation of an enhanced IHE architecture demonstrating how ebXML Business Processes, IHE Workflow Profiles and the IHE XDS architecture can all be integrated to provide collaborative business process support in the healthcare domain.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"6 1","pages":"91-91"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75104754","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 applications of biometric databases the typical task is to identify individuals according to features which are not exactly known. Reasons for this inexactness are varying measuring techniques or environmental circumstances. Since these circumstances are not necessarily the same when determining the features for different individuals, the exactness might strongly vary between the individuals as well as between the features. To identify individuals, similarity search on feature vectors is applicable, but even the use of adaptable distance measures is not capable to handle objects having an individual level of exactness. Therefore, we develop a comprehensive probabilistic theory in which uncertain observations are modeled by probabilistic feature vectors (pfv), i.e. feature vectors where the conventional feature values are replaced by Gaussian probability distribution functions. Each feature value of each object is complemented by a variance value indicating its uncertainty. We define two types of identification queries, k-mostlikely identification and threshold identification. For efficient query processing, we propose a novel index structure, the Gauss-tree. Our experimental evaluation demonstrates that pfv stored in a Gauss-tree significantly improve the result quality compared to traditional feature vectors. Additionally, we show that the Gauss-tree significantly speeds up query times compared to competitive methods.
{"title":"The Gauss-Tree: Efficient Object Identification in Databases of Probabilistic Feature Vectors","authors":"C. Böhm, A. Pryakhin, Matthias Schubert","doi":"10.1109/ICDE.2006.159","DOIUrl":"https://doi.org/10.1109/ICDE.2006.159","url":null,"abstract":"In applications of biometric databases the typical task is to identify individuals according to features which are not exactly known. Reasons for this inexactness are varying measuring techniques or environmental circumstances. Since these circumstances are not necessarily the same when determining the features for different individuals, the exactness might strongly vary between the individuals as well as between the features. To identify individuals, similarity search on feature vectors is applicable, but even the use of adaptable distance measures is not capable to handle objects having an individual level of exactness. Therefore, we develop a comprehensive probabilistic theory in which uncertain observations are modeled by probabilistic feature vectors (pfv), i.e. feature vectors where the conventional feature values are replaced by Gaussian probability distribution functions. Each feature value of each object is complemented by a variance value indicating its uncertainty. We define two types of identification queries, k-mostlikely identification and threshold identification. For efficient query processing, we propose a novel index structure, the Gauss-tree. Our experimental evaluation demonstrates that pfv stored in a Gauss-tree significantly improve the result quality compared to traditional feature vectors. Additionally, we show that the Gauss-tree significantly speeds up query times compared to competitive methods.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"11 1","pages":"9-9"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73792085","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}
Our society is more dependent on information systems than ever before. However, managing the information systems infrastructure in a cost-effective manner is a growing challenge. The total cost of ownership (TCO) of information technology is increasingly dominated by people costs. In fact, mistakes in operations and administration of information systems are the single most reasons for system outage and unacceptable performance. For information systems to provide value to their customers, we must reduce the complexity associated with their deployment and usage.
{"title":"Foundations of Automated Database Tuning","authors":"Surajit Chaudhuri, G. Weikum","doi":"10.1109/ICDE.2006.72","DOIUrl":"https://doi.org/10.1109/ICDE.2006.72","url":null,"abstract":"Our society is more dependent on information systems than ever before. However, managing the information systems infrastructure in a cost-effective manner is a growing challenge. The total cost of ownership (TCO) of information technology is increasingly dominated by people costs. In fact, mistakes in operations and administration of information systems are the single most reasons for system outage and unacceptable performance. For information systems to provide value to their customers, we must reduce the complexity associated with their deployment and usage.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"1 1","pages":"104-104"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91297915","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}
P. Pietzuch, J. Ledlie, Jeffrey Shneidman, M. Roussopoulos, M. Welsh, M. Seltzer
To use their pool of resources efficiently, distributed stream-processing systems push query operators to nodes within the network. Currently, these operators, ranging from simple filters to custom business logic, are placed manually at intermediate nodes along the transmission path to meet application-specific performance goals. Determining placement locations is challenging because network and node conditions change over time and because streams may interact with each other, opening venues for reuse and repositioning of operators. This paper describes a stream-based overlay network (SBON), a layer between a stream-processing system and the physical network that manages operator placement for stream-processing systems. Our design is based on a cost space, an abstract representation of the network and on-going streams, which permits decentralized, large-scale multi-query optimization decisions. We present an evaluation of the SBON approach through simulation, experiments on PlanetLab, and an integration with Borealis, an existing stream-processing engine. Our results show that an SBON consistently improves network utilization, provides low stream latency, and enables dynamic optimization at low engineering cost.
{"title":"Network-Aware Operator Placement for Stream-Processing Systems","authors":"P. Pietzuch, J. Ledlie, Jeffrey Shneidman, M. Roussopoulos, M. Welsh, M. Seltzer","doi":"10.1109/ICDE.2006.105","DOIUrl":"https://doi.org/10.1109/ICDE.2006.105","url":null,"abstract":"To use their pool of resources efficiently, distributed stream-processing systems push query operators to nodes within the network. Currently, these operators, ranging from simple filters to custom business logic, are placed manually at intermediate nodes along the transmission path to meet application-specific performance goals. Determining placement locations is challenging because network and node conditions change over time and because streams may interact with each other, opening venues for reuse and repositioning of operators. This paper describes a stream-based overlay network (SBON), a layer between a stream-processing system and the physical network that manages operator placement for stream-processing systems. Our design is based on a cost space, an abstract representation of the network and on-going streams, which permits decentralized, large-scale multi-query optimization decisions. We present an evaluation of the SBON approach through simulation, experiments on PlanetLab, and an integration with Borealis, an existing stream-processing engine. Our results show that an SBON consistently improves network utilization, provides low stream latency, and enables dynamic optimization at low engineering cost.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"122 1","pages":"49-49"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83649866","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}
Todays abundance of storage coupled with digital technologies in virtually any scientific or commercial application such as medical and biological imaging or music archives deal with tremendous quantities of images, videos or audio files stored in large multimedia databases. For content-based data mining and retrieval purposes suitable similarity models are crucial. The Earth Mover’s Distance was introduced in Computer Vision to better approach human perceptual similarities. Its computation, however, is too complex for usage in interactive multimedia database scenarios. In order to enable efficient query processing in large databases, we propose an index-supported multistep algorithm. We therefore develop new lower bounding approximation techniques for the Earth Mover’s Distance which satisfy high quality criteria including completeness (no false drops), index-suitability and fast computation. We demonstrate the efficiency of our approach in extensive experiments on large image databases
{"title":"Approximation Techniques for Indexing the Earth Mover’s Distance in Multimedia Databases","authors":"I. Assent, Andrea Wenning, T. Seidl","doi":"10.1109/ICDE.2006.25","DOIUrl":"https://doi.org/10.1109/ICDE.2006.25","url":null,"abstract":"Todays abundance of storage coupled with digital technologies in virtually any scientific or commercial application such as medical and biological imaging or music archives deal with tremendous quantities of images, videos or audio files stored in large multimedia databases. For content-based data mining and retrieval purposes suitable similarity models are crucial. The Earth Mover’s Distance was introduced in Computer Vision to better approach human perceptual similarities. Its computation, however, is too complex for usage in interactive multimedia database scenarios. In order to enable efficient query processing in large databases, we propose an index-supported multistep algorithm. We therefore develop new lower bounding approximation techniques for the Earth Mover’s Distance which satisfy high quality criteria including completeness (no false drops), index-suitability and fast computation. We demonstrate the efficiency of our approach in extensive experiments on large image databases","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"91 1","pages":"11-11"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83790023","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}
Data mining promises to discover valid and potentially useful patterns in data. Often, discovered patterns are not useful to the user."Actionability" addresses this problem in that a pattern is deemed actionable if the user can act upon it in her favor. We introduce the notion of "action" as a domain-independent way to model the domain knowledge. Given a data set about actionable features and an utility measure, a pattern is actionable if it summarizes a population that can be acted upon towards a more promising population observed with a higher utility. We present several pruning strategies taking into account the actionability requirement to reduce the search space, and algorithms for mining all actionable patterns as well as mining the top k actionable patterns. We evaluate the usefulness of patterns and the focus of search on a real-world application domain.
{"title":"Mining Actionable Patterns by Role Models","authors":"Ke Wang, Yuelong Jiang, A. Tuzhilin","doi":"10.1109/ICDE.2006.96","DOIUrl":"https://doi.org/10.1109/ICDE.2006.96","url":null,"abstract":"Data mining promises to discover valid and potentially useful patterns in data. Often, discovered patterns are not useful to the user.\"Actionability\" addresses this problem in that a pattern is deemed actionable if the user can act upon it in her favor. We introduce the notion of \"action\" as a domain-independent way to model the domain knowledge. Given a data set about actionable features and an utility measure, a pattern is actionable if it summarizes a population that can be acted upon towards a more promising population observed with a higher utility. We present several pruning strategies taking into account the actionability requirement to reduce the search space, and algorithms for mining all actionable patterns as well as mining the top k actionable patterns. We evaluate the usefulness of patterns and the focus of search on a real-world application domain.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"2 1","pages":"16-16"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84582011","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}
Ganesh Ramakrishnan, Sachindra Joshi, Sumit Negi, R. Krishnapuram, S. Balakrishnan
Speed to market is critical to companies that are driven by sales in a competitive market. The earlier a potential customer can be approached in the decision making process of a purchase, the higher are the chances of converting that prospect into a customer. Traditional methods to identify sales leads such as company surveys and direct marketing are manual, expensive and not scalable. Over the past decade the World Wide Web has grown into an information-mesh, with most important facts being reported through Web sites. Several news papers, press releases, trade journals, business magazines and other related sources are on-line. These sources could be used to identify prospective buyers automatically. In this paper, we present a system called ETAP (Electronic Trigger Alert Program) that extracts trigger events from Web data that help in identifying prospective buyers. Trigger events are events of corporate relevance and indicative of the propensity of companies to purchase new products associated with these events. Examples of trigger events are change in management, revenue growth and mergers & acquisitions. The unstructured nature of information makes the extraction task of trigger events difficult. We pose the problem of trigger events extraction as a classification problem and develop methods for learning trigger event classifiers using existing classification methods. We present methods to automatically generate the training data required to learn the classifiers. We also propose a method of feature abstraction that uses named entity recognition to solve the problem of data sparsity. We score and rank the trigger events extracted from ETAP for easy browsing. Our experiments show the effectiveness of the method and thus establish the feasibility of automatic sales lead generation using the Web data.
{"title":"Automatic Sales Lead Generation from Web Data","authors":"Ganesh Ramakrishnan, Sachindra Joshi, Sumit Negi, R. Krishnapuram, S. Balakrishnan","doi":"10.1109/ICDE.2006.28","DOIUrl":"https://doi.org/10.1109/ICDE.2006.28","url":null,"abstract":"Speed to market is critical to companies that are driven by sales in a competitive market. The earlier a potential customer can be approached in the decision making process of a purchase, the higher are the chances of converting that prospect into a customer. Traditional methods to identify sales leads such as company surveys and direct marketing are manual, expensive and not scalable. Over the past decade the World Wide Web has grown into an information-mesh, with most important facts being reported through Web sites. Several news papers, press releases, trade journals, business magazines and other related sources are on-line. These sources could be used to identify prospective buyers automatically. In this paper, we present a system called ETAP (Electronic Trigger Alert Program) that extracts trigger events from Web data that help in identifying prospective buyers. Trigger events are events of corporate relevance and indicative of the propensity of companies to purchase new products associated with these events. Examples of trigger events are change in management, revenue growth and mergers & acquisitions. The unstructured nature of information makes the extraction task of trigger events difficult. We pose the problem of trigger events extraction as a classification problem and develop methods for learning trigger event classifiers using existing classification methods. We present methods to automatically generate the training data required to learn the classifiers. We also propose a method of feature abstraction that uses named entity recognition to solve the problem of data sparsity. We score and rank the trigger events extracted from ETAP for easy browsing. Our experiments show the effectiveness of the method and thus establish the feasibility of automatic sales lead generation using the Web data.","PeriodicalId":6819,"journal":{"name":"22nd International Conference on Data Engineering (ICDE'06)","volume":"67 1","pages":"101-101"},"PeriodicalIF":0.0,"publicationDate":"2006-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90276794","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：