Meeting of the Association for Computational Linguistics, 2008
This paper describes a computational ap- proach to resolving the true referent of a named mention... more This paper describes a computational ap- proach to resolving the true referent of a named mention of a person in the body of an email. A generative model of mention gener- ation is used to guide mention resolution. Re- sults on three relatively small collections indi- cate that the accuracy of this approach com- pares favorably to the best known
Access to historically significant email archives poses challenges that arise less often in perso... more Access to historically significant email archives poses challenges that arise less often in personal collections. Most notably, searchers may need help making sense of the identities, roles, and relationships of individuals that participated in archived email exchanges. This paper describes an exploratory study of identity resolution in the public subset of the Enron collection. Address- name and address-address associations in
As the Internet currently represents the dominating information resource, the need for more relia... more As the Internet currently represents the dominating information resource, the need for more reliable web search engines gets increasing research focus. While many investigations have explored the use of implicit feedback to improve the user search results, no study has yet examined a more direct approach of the user's mental state disclosure, his facial expressions while examining these results. This paper tries to answer the following question: can we predict document relevance from the user's facial expressions? The intuition is that the user mental state reflected in his facial expression while examining a relevant document is different from his expression while examining a non-relevant one. A conducted feasibility study revealed that the detected facial expressions corresponding to relevant and non-relevant documents were distinguishable by a trained neural network classifier but many collected data are required to predict the relevance of future documents.
A topic tracking system that combines elements from vector space and language modeling frameworks... more A topic tracking system that combines elements from vector space and language modeling frameworks to compute document scores is described. The model is used for both the traditional TDT topic tracking evaluation design and the new supervised adaptive topic tracking evaluation. Results indicate that supervised adaptation and score normalization should be more closely coupled, and that current techniques for detection error tradeoff analysis may be of limited utility when supervised adaptation is performed.
Automatic knowledge base population from text is an important technology for a broad range of app... more Automatic knowledge base population from text is an important technology for a broad range of approaches to learning by reading. Effective automated knowledge base population depends critically upon coreference resolution of entities across sources. Use of a wide range of features, both those that capture evidence for entity merging and those that argue against merging, can significantly improve machine learning-based cross-document coreference resolution.
In this work we consider the design principles of the Instance-Based Netwo rk (IBN), an extended ... more In this work we consider the design principles of the Instance-Based Netwo rk (IBN), an extended version of a generic Content-Based Network (CBN). IBN acts as an overlay com- munication platform over which end-point entities, called contents, communic ate indepen- dently from their physical locations while providing the flexibility of having differ ent in- stances of the same content. The semantics of different instances are as signed by the appli- cation using the IBN. Routing in the IBN is instance-based; the IBN can route a message to a specific content instance or to the closest instance, if no exact match isfound for the destination content instance.
Meeting of the Association for Computational Linguistics, 2008
This paper describes a computational ap- proach to resolving the true referent of a named mention... more This paper describes a computational ap- proach to resolving the true referent of a named mention of a person in the body of an email. A generative model of mention gener- ation is used to guide mention resolution. Re- sults on three relatively small collections indi- cate that the accuracy of this approach com- pares favorably to the best known
Access to historically significant email archives poses challenges that arise less often in perso... more Access to historically significant email archives poses challenges that arise less often in personal collections. Most notably, searchers may need help making sense of the identities, roles, and relationships of individuals that participated in archived email exchanges. This paper describes an exploratory study of identity resolution in the public subset of the Enron collection. Address- name and address-address associations in
As the Internet currently represents the dominating information resource, the need for more relia... more As the Internet currently represents the dominating information resource, the need for more reliable web search engines gets increasing research focus. While many investigations have explored the use of implicit feedback to improve the user search results, no study has yet examined a more direct approach of the user's mental state disclosure, his facial expressions while examining these results. This paper tries to answer the following question: can we predict document relevance from the user's facial expressions? The intuition is that the user mental state reflected in his facial expression while examining a relevant document is different from his expression while examining a non-relevant one. A conducted feasibility study revealed that the detected facial expressions corresponding to relevant and non-relevant documents were distinguishable by a trained neural network classifier but many collected data are required to predict the relevance of future documents.
A topic tracking system that combines elements from vector space and language modeling frameworks... more A topic tracking system that combines elements from vector space and language modeling frameworks to compute document scores is described. The model is used for both the traditional TDT topic tracking evaluation design and the new supervised adaptive topic tracking evaluation. Results indicate that supervised adaptation and score normalization should be more closely coupled, and that current techniques for detection error tradeoff analysis may be of limited utility when supervised adaptation is performed.
Automatic knowledge base population from text is an important technology for a broad range of app... more Automatic knowledge base population from text is an important technology for a broad range of approaches to learning by reading. Effective automated knowledge base population depends critically upon coreference resolution of entities across sources. Use of a wide range of features, both those that capture evidence for entity merging and those that argue against merging, can significantly improve machine learning-based cross-document coreference resolution.
In this work we consider the design principles of the Instance-Based Netwo rk (IBN), an extended ... more In this work we consider the design principles of the Instance-Based Netwo rk (IBN), an extended version of a generic Content-Based Network (CBN). IBN acts as an overlay com- munication platform over which end-point entities, called contents, communic ate indepen- dently from their physical locations while providing the flexibility of having differ ent in- stances of the same content. The semantics of different instances are as signed by the appli- cation using the IBN. Routing in the IBN is instance-based; the IBN can route a message to a specific content instance or to the closest instance, if no exact match isfound for the destination content instance.
Uploads
Papers by Tamer Elsayed