�ݺ�ߣshows by User: GraceYang22

�ݺ�ߣshows by User: GraceYang22 / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: GraceYang22 / Mon, 25 Apr 2016 14:53:08 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: GraceYang22 Tips for writing a paper /slideshow/tips-for-writing-a-paper/61326324 tipsforwritingapaper-160425145308
This is a summary of the writing tips that I always nag to the students. The slides were created by my student Hongkai Wu, while he was struggling to write his Master's thesis. ]]>
This is a summary of the writing tips that I always nag to the students. The slides were created by my student Hongkai Wu, while he was struggling to write his Master's thesis. ]]> Mon, 25 Apr 2016 14:53:08 GMT /slideshow/tips-for-writing-a-paper/61326324 GraceYang22@slideshare.net(GraceYang22) Tips for writing a paper GraceYang22 This is a summary of the writing tips that I always nag to the students. The slides were created by my student Hongkai Wu, while he was struggling to write his Master's thesis. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/tipsforwritingapaper-160425145308-thumbnail.jpg?width=120&height=120&fit=bounds" /> This is a summary of the writing tips that I always nag to the students. The slides were created by my student Hongkai Wu, while he was struggling to write his Master's thesis.

Tips for writing a paper from Grace Yang

]]> 2593 11 https://cdn.slidesharecdn.com/ss_thumbnails/tipsforwritingapaper-160425145308-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Win-Win Search: Dual-Agent Stochastic Game in Session Search (SIGIR 2014) /slideshow/winwin-search-dualagent-stochastic-game-in-session-search-sigir-2014/53477352 sigir-2014-full-session-v10-short-151002200637-lva1-app6891
The POMDP modeling of dynamic search. It is our second model for dynamic search. The first POMDP model for dynamic search. ]]>
The POMDP modeling of dynamic search. It is our second model for dynamic search. The first POMDP model for dynamic search. ]]> Fri, 02 Oct 2015 20:06:37 GMT /slideshow/winwin-search-dualagent-stochastic-game-in-session-search-sigir-2014/53477352 GraceYang22@slideshare.net(GraceYang22) Win-Win Search: Dual-Agent Stochastic Game in Session Search (SIGIR 2014) GraceYang22 The POMDP modeling of dynamic search. It is our second model for dynamic search. The first POMDP model for dynamic search. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/sigir-2014-full-session-v10-short-151002200637-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds" /> The POMDP modeling of dynamic search. It is our second model for dynamic search. The first POMDP model for dynamic search.

Win-Win Search: Dual-Agent Stochastic Game in Session Search (SIGIR 2014) from Grace Yang

]]> 1232 12 https://cdn.slidesharecdn.com/ss_thumbnails/sigir-2014-full-session-v10-short-151002200637-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 The Water Filling Model and The Cube Test: Multi-Dimensional Evaluation for Professional Search (CIKM 2013) /slideshow/the-water-filling-model-and-the-cube-test-multidimensional-evaluation-for-professional-search-cikm-2013/53476950 cikm-7-151002195242-lva1-app6891
This talk describes the Cube Test, an IR evaluation metric that we create for measuring system effectiveness for the entire information seeking process. Featuring modeling of a cap to stop (task completion), multiple subtopics, time aspect, subtopic importance, subtopic relatedness, volume of gain. ]]>
This talk describes the Cube Test, an IR evaluation metric that we create for measuring system effectiveness for the entire information seeking process. Featuring modeling of a cap to stop (task completion), multiple subtopics, time aspect, subtopic importance, subtopic relatedness, volume of gain. ]]> Fri, 02 Oct 2015 19:52:41 GMT /slideshow/the-water-filling-model-and-the-cube-test-multidimensional-evaluation-for-professional-search-cikm-2013/53476950 GraceYang22@slideshare.net(GraceYang22) The Water Filling Model and The Cube Test: Multi-Dimensional Evaluation for Professional Search (CIKM 2013) GraceYang22 This talk describes the Cube Test, an IR evaluation metric that we create for measuring system effectiveness for the entire information seeking process. Featuring modeling of a cap to stop (task completion), multiple subtopics, time aspect, subtopic importance, subtopic relatedness, volume of gain. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/cikm-7-151002195242-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds" /> This talk describes the Cube Test, an IR evaluation metric that we create for measuring system effectiveness for the entire information seeking process. Featuring modeling of a cap to stop (task completion), multiple subtopics, time aspect, subtopic importance, subtopic relatedness, volume of gain.

The Water Filling Model and The Cube Test: Multi-Dimensional Evaluation for Professional Search (CIKM 2013) from Grace Yang

]]> 938 9 https://cdn.slidesharecdn.com/ss_thumbnails/cikm-7-151002195242-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Utilizing Query Change for Session Search (SIGIR 2013) /slideshow/utilizing-query-change-for-session-search-sigir-2013/53475165 sigir2013-full-ver8-151002185153-lva1-app6892
Query Change Model (QCM) for dynamic search. The Markov Decision Model. Our first model for dynamic search. ]]>
Query Change Model (QCM) for dynamic search. The Markov Decision Model. Our first model for dynamic search. ]]> Fri, 02 Oct 2015 18:51:53 GMT /slideshow/utilizing-query-change-for-session-search-sigir-2013/53475165 GraceYang22@slideshare.net(GraceYang22) Utilizing Query Change for Session Search (SIGIR 2013) GraceYang22 Query Change Model (QCM) for dynamic search. The Markov Decision Model. Our first model for dynamic search. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/sigir2013-full-ver8-151002185153-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds" /> Query Change Model (QCM) for dynamic search. The Markov Decision Model. Our first model for dynamic search.

Utilizing Query Change for Session Search (SIGIR 2013) from Grace Yang

]]> 724 7 https://cdn.slidesharecdn.com/ss_thumbnails/sigir2013-full-ver8-151002185153-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Designing States, Actions, and Rewards for Using POMDP in Session Search /slideshow/designing-states-actions-and-rewards-for-using-pomdp-in-session-search/53474786 ecir-2015-full-session-v8-151002183757-lva1-app6891
Coming out with the states, actions, rewards design for an application is an art. We discuss among the available options in IR and evaluate the options' effectiveness and efficiency. ]]>
Coming out with the states, actions, rewards design for an application is an art. We discuss among the available options in IR and evaluate the options' effectiveness and efficiency. ]]> Fri, 02 Oct 2015 18:37:57 GMT /slideshow/designing-states-actions-and-rewards-for-using-pomdp-in-session-search/53474786 GraceYang22@slideshare.net(GraceYang22) Designing States, Actions, and Rewards for Using POMDP in Session Search GraceYang22 Coming out with the states, actions, rewards design for an application is an art. We discuss among the available options in IR and evaluate the options' effectiveness and efficiency. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/ecir-2015-full-session-v8-151002183757-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds" /> Coming out with the states, actions, rewards design for an application is an art. We discuss among the available options in IR and evaluate the options' effectiveness and efficiency.

Designing States, Actions, and Rewards for Using POMDP in Session Search from Grace Yang

]]> 827 9 https://cdn.slidesharecdn.com/ss_thumbnails/ecir-2015-full-session-v8-151002183757-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Learning to Reinforce Search Effectiveness /slideshow/learning-to-reinforce-search-effectiveness/53474467 em-151002182642-lva1-app6892
We use contextual bandit and EM for modeling the two communication between the user and the search engine. The 4th algorithm that we've created for dynamic search]]>
We use contextual bandit and EM for modeling the two communication between the user and the search engine. The 4th algorithm that we've created for dynamic search]]> Fri, 02 Oct 2015 18:26:42 GMT /slideshow/learning-to-reinforce-search-effectiveness/53474467 GraceYang22@slideshare.net(GraceYang22) Learning to Reinforce Search Effectiveness GraceYang22 We use contextual bandit and EM for modeling the two communication between the user and the search engine. The 4th algorithm that we've created for dynamic search <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/em-151002182642-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds" /> We use contextual bandit and EM for modeling the two communication between the user and the search engine. The 4th algorithm that we've created for dynamic search

Learning to Reinforce Search Effectiveness from Grace Yang

]]> 826 7 https://cdn.slidesharecdn.com/ss_thumbnails/em-151002182642-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Session Search by Direct Policy Learning /slideshow/session-search-by-direct-policy-learning/53472916 dpl-151002174020-lva1-app6891
Modeling Information Seeking by Direct Policy Learning. Our 3rd model for dynamic IR. Presented at ICTIR 2015. ]]>
Modeling Information Seeking by Direct Policy Learning. Our 3rd model for dynamic IR. Presented at ICTIR 2015. ]]> Fri, 02 Oct 2015 17:40:20 GMT /slideshow/session-search-by-direct-policy-learning/53472916 GraceYang22@slideshare.net(GraceYang22) Session Search by Direct Policy Learning GraceYang22 Modeling Information Seeking by Direct Policy Learning. Our 3rd model for dynamic IR. Presented at ICTIR 2015. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/dpl-151002174020-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds" /> Modeling Information Seeking by Direct Policy Learning. Our 3rd model for dynamic IR. Presented at ICTIR 2015.

Session Search by Direct Policy Learning from Grace Yang

]]> 870 9 https://cdn.slidesharecdn.com/ss_thumbnails/dpl-151002174020-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 https://public.slidesharecdn.com/v2/images/profile-picture.png Professor/Researcher in Information Retrieval. infosense.cs.georgetown.edu/grace/ https://cdn.slidesharecdn.com/ss_thumbnails/tipsforwritingapaper-160425145308-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/tips-for-writing-a-paper/61326324 Tips for writing a paper https://cdn.slidesharecdn.com/ss_thumbnails/sigir-2014-full-session-v10-short-151002200637-lva1-app6891-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/winwin-search-dualagent-stochastic-game-in-session-search-sigir-2014/53477352 Win-Win Search: Dual-A... https://cdn.slidesharecdn.com/ss_thumbnails/cikm-7-151002195242-lva1-app6891-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/the-water-filling-model-and-the-cube-test-multidimensional-evaluation-for-professional-search-cikm-2013/53476950 The Water Filling Mode...