Videos with tag data
Results 1-8 of 11
 
00:31.44
00:31.44
00:31.44

Next Stop Recommender-30-second video

User wandering behaviours may involve many location visits in different order. The research team has proposed an algorithm which can provide users recommendation for their next visit according to the similarity of their behaviours between each others and the connections amongst locations. In order to test the effectiveness of proposed algorithm the research team develops a mobile app Next Stop Recommender for Android platform. The mobile app and the web user interface can collect sequential positioning data of travelers. The server side Java servlets can find the visit lists of travelers and summarize popular route patterns; moreover, it can turn the route patterns into recommendation rules and make recommendations for travelers. We can record mobile learner's behaviors (including where they spent more time and what is the sequence for learning) and compare one's with others to make recommendations (could be more than one recommendation when the learner asks) for the mobile learners.

Added: 1456 days ago by maiga

Views: 538 | Comments: 0 | Not yet rated

 
13:15
13:15
13:15

Graph data visualization and layout using game physics

This research report proposes a new approach and a prototype for a new method for visualization that makes the user the central factor in the layout of a graph diagram. This method proposes to use a game physics engine to create a prototype that can provide realistic behaviour to objects and allow users to create their own graph layouts in seconds without any layout algorithm other than the original configuration of fixture types for a physics engine. A technology framework and tools were selected and a prototype was constructed to evaluate the approach and the method was proven to be viable and the prototype was deployed. Future research will be required to improve graphic presentation and allow other datasets to be visualized in combination with more interaction capabilities that allow the user to explore the data and discover new insights more easily than in the past.

Added: 1515 days ago by GlennIrwin

Views: 296 | Comments: 0 | Not yet rated

 
07:04
07:04
07:04

Next Stop Recommendation Service

Next Stop Recommendation Service uses time-series data mining technique to calculate next stop recommendation according to users' wandering behaviours. Dirksen Liu and Maiga Chang. (2011). Next-Stop Recommendation to Travelers according to Their Sequential Wandering Behaviours. Journal of Internet Technology, 12(1), 171-179. (SCI-E, impact factor 0.448) http://nextstop.dyndns.info:4735

Added: 1620 days ago by maiga

Views: 305 | Comments: 0 | Not yet rated

 
18:36
18:36
18:36

Goal-Oriented Requirement Analysis for Data Warehouse Design

Alex Lau TMA1

Added: 2419 days ago by shulau

Views: 509 | Comments: 0 | Not yet rated

 
19:11
19:11
19:11

Taking account of privacy when designing cloud computing ser

http://0-delivery.acm.org.aupac.lib.athabascau.ca/10.1145/1570000/1564628/05071532.pdf?key1=1564628&key2=1444135921&coll=DL&dl=ACM&CFID=6300359&CFTOKEN=33574409 This paper appears in: CLOUD '09. ICSE Workshop on Software Engineering Challenges of Cloud Computing Issue Date: 23rd May 2009 On page(s): 44 - 52 Location: Vancouver, BC Print ISBN: 978-1-4244-3713-9 INSPEC Accession Number: 10718493 Digital Object Identifier: 10.1109/CLOUD.2009.5071532  Date of Current Version: 16 June 2009

Added: 2432 days ago by 2743924

Views: 307 | Comments: 0 | Not yet rated

 
19:29
19:29
19:29

Updated Presentation - Using Formal Specification Models and

Many people depend on web applications for important daily activities. But web application quality has not kept pace with popularity. Hence, improved web application quality is needed and increased testing can help. One step in this direction involves creating schemes to support automatic generation of navigation test cases. Approaches involving formal models, Unified Modeling Language development models and other mechanisms have been tried. But they are always used in isolation. Consequently, a combination of formal specification models and UML models for the generation of test cases is proposed. To evaluate the effectiveness of this approach a case study is used. Combined model based test cases are compared to captured user sessions of volunteers. Since test cases and user sessions are just sequences of logical web pages, Levenshtein distance is used for comparison purposes. Preliminary empirical results show there is value to the approach in that the combined test set does a better job of mimicking real user sessions than does test cases from UML models alone. There are a few advantages to the approach. It is easy to create the models and subsequent test case generation can be completely automated. With the approach, test cases are available very early in the development process when it is more economical to repair the problems they identify. Contributions made from the study involve the proposal of an innovative model combination, evidence to support its merit and the identification of implementation hurdles.

Added: 2614 days ago by johnka15

Views: 384 | Comments: 0 | Not yet rated

 
19:29
19:29
19:29

Using Formal Specification Models and Early Development Mode

Many people depend on web applications for important daily activities. But web application quality has not kept pace with popularity. Hence, improved web application quality is needed and increased testing can help. One step in this direction involves creating schemes to support automatic generation of navigation test cases. Approaches involving formal models, Unified Modeling Language development models and other mechanisms have been tried. But they are always used in isolation. Consequently, a combination of formal specification models and UML models for the generation of test cases is proposed. To evaluate the effectiveness of this approach a case study is used. Combined model based test cases are compared to captured user sessions of volunteers. Since test cases and user sessions are just sequences of logical web pages, Levenshtein distance is used for comparison purposes. Preliminary empirical results show there is value to the approach in that the combined test set does a better job of mimicking real user sessions than does test cases from UML models alone. There are a few advantages to the approach. It is easy to create the models and subsequent test case generation can be completely automated. With the approach, test cases are available very early in the development process when it is more economical to repair the problems they identify. Contributions made from the study involve the proposal of an innovative model combination, evidence to support its merit and the identification of implementation hurdles

Added: 2614 days ago by johnka15

Views: 472 | Comments: 0 | Not yet rated

 
26:28
26:28
26:28

Classifying Software Changes: Clean or Buggy?

Abstract: The paper under review suggests novel technique, Change Classification, for predicting latent software bugs by implementing machine learning algorithms, namely the Support Vector Machine algorithm to classify changes as either clean or buggy immediately after the commit has been made. The classifier is trained using features in the SCM repository of the system under investigation. Results were tested with 12 open source projects with 78% accuracy and a 60% buggy change recall on average.

Added: 3044 days ago by mohammedy

Views: 268 | Comments: 0 | Not yet rated