Search

Segmentation methods used in marketing literature range from those based on demographic variables to those relying on more sophisticated clustering techniques. Grouping consumers into relatively homogeneous classes is of great importance for marketers since it enables them to develop specific products and marketing strategies to better cater these segments. A new segmentation method based on the heterogeneity among consumers' perceptions of underlying product-related factors, is proposed. Specifically, variation of factor loadings, coupled with a highly intensive computer technique (bootstrap) is used to classify consumers into homogeneous groups. The validity of this segmentation scheme is compared to those of traditional segmentation methods, by examining the forecasting abilities of a choice model under these different scenarios.

Technical report TR95-17. Information Superhighway environments such as the Internet have brough us ready access to large amounts of information. However, Internet data is notoriously unorganized and autonomously managed in a distributed fashion. Large scale information monitoring in the Internet environment requires support beyond traditional database techniques. Two of the key issues are the increasing reward in monitoring a fast growing information base and the similarly increasing processing cost. To improve the expressiveness of queries for information monitoring, we define continual queries as a useful tool for monitoring of updated information. Continual queries are standing queries that monitor the source data and notify the users whenever new data matches the query. In addition to periodic refresh, continual queries include Epsilon Transaction concepts to allow users to specify query refresh based on the magnitude of updates. To support effiecient processing of continual queries, we propose a differential re-evaluation algorithm (DRA), which exploits the structure and information contained in both the query expressions and the database update operations. The DRA design can be seen as a synthesis of previous research on differential files, incrementala view maintenance, and active databases. | TRID-ID TR95-17

The objective of this study is to evaluate a neural network monitoring system for continuous surveillance of the NASA Space Station Freedom, to detect and locate atmospheric leakage. The monitoring system uses surface mounted sensors to detect acoustic emission signals produced by the leak, which are then relayed to the neural network for source location determination. To date, acoustic emission leak location systems have achieved only limited success and are adversely affected by noise and complex geometries. For a monitoring system to be effective in locating atmospheric leakage it must have the ability to mask out noise, circumvent multipath interference, and process AE signals in real time. Neural networks seem ideally suited to the problem.© (1991) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Neural networks were applied to the classification oftwo types ofacoustic emission (AE) events crack growth andfretting a simulated aiiframejoint specimen. Signals were obtainedfromfour sensors at different locations on the test specimen. Multilayered neural networks were trained to classify the signals using the error backpropagation learning algorithm enabling AE events arisingfrom crack growth to be distinguishedfrom those caused by fretting. In thispaper we evaluate the neural network classWcationperformancefor sensor location dependent and sensor location independent training and testing sets. Further we present a new training strategy which signcantly reduces the time required to learn large training sets using the error backpropagation learning algorithm and improves the generalization performance of the network. 1.© (1990) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Several studies have recently concentrated on the generation of wrappers for extracting data from Web data sources. The ROADRUNNER system aims at automating the tedious and expensive process of writing wrappers in an unsupervised, domain-independent, and scalable manner. The system is based on a grammar inference algorithm, called MATCH, which has been designed in a sound theoretical framework. However, in its original definition MATCH lacks in expressivity; that is, in many cases when MATCH runs over real-life Web pages, it is not able to produce a solution. In this paper we address the challenging issue of developing techniques that allow us to build upon MATCH an effective and efficient system, without renouncing to the original formal background. First, we analyze the main limitations of MATCH; then we illustrate the techniques we have developed to overcome such limitations. Finally we report on the results of some experiments, that show the efficacy of the introduced techniques and demonstrate the improvements of the overall system.