User Tools

Site Tools


projects:year4:15.2

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Next revision
Previous revision
projects:year4:15.2 [2019/08/21 11:10]
sally.johnson created
projects:year4:15.2 [2021/06/02 15:25] (current)
sally.johnson [Table]
Line 4: Line 4:
 **Objectives:** **Objectives:**
  
-The goal of this project is to develop a multi-modal retrieval system that can work with both image and textual information for question answering. The project creates a unified, graph-based representation model in both text and image domains and is capable of using visual features to search in a textual database or use textual queries to retrieve relevant images from a database. +The project aims to develop and evaluate robust link predicting methods incorporating ensemble and incremental learning capability. The objectives are to: 
- +  * Implement multiple link prediction methods 
-To accomplish this, the project aims to develop a computational toolbox for automatically generating graph representation of images and text information, and an API set for interacting with a database consisting of graph-based information. +  * Develop and evaluate the ensemble link prediction methods 
- +  * Develop an incremental learning method and evaluate the performance of the model 
-To achieve this, following action items were considered+  * Design and implement an interactive visualization system
-  * Build a multi-scale feature extraction from images and capture the spatial relationship of image features in the form of a graph. +
-  * Incorporate the dependency graphs obtained from textual information annotated with syntactic information and store the resulting data in the form of a graph database. +
-  * Develop a unifying platform for correlating images and textual information represented in terms of multi-scale image features and dependency text representation. +
-  * Develop retrieval tools for a question answering system that can support a question answering system between both modalities. +
-  * Build a proof of concept API for the system.+
  
 </WRAP> </WRAP>
 ===== Project - Team ===== ===== Project - Team =====
-^ Team Member        ^ Role         ^ Email          ^ Phone Number   ^ Academic Site/IAB  +^ Team Member           ^ Role              ^ Email                   ^ Phone Number    ^ Academic Site/IAB            
-Ali Shokoufandeh   | PI           | Not available  | Not available  Drexel University +Vijay Raghavan        | PI                | raghavan@louisiana.edu  | (337) 482-6603  | UL Lafayette                 | 
-Yusuf Osmanlioglu  PhD Student  | Not available  | Not available Drexel University  |+| Raju Gottumukkala     | PI                | raju@louisiana.edu      | (337) 482-0632  | UL Lafayette                 | 
 +| Ryan Benton           | PI                | Not available           | Not available   | University of South Alabama  | 
 +Murali K. Pusala      Graduate Student  | Not available           | Not available   | UL Lafayette                 | 
 +| Amirhossein Tavanaei  | Graduate Student  | Not available           | Not available   | UL Lafayette                 | 
 +| Narendra Sanikommu    | Graduate Student  Not available           | Not available   | UL Lafayette                 | 
 +| Jaya Krishna          | Graduate Student  | Not available           | Not available   | UL Lafayette                 |
  
  
Line 25: Line 25:
 ===== Project - Impact and Uses/Benefits ===== ===== Project - Impact and Uses/Benefits =====
 <WRAP leftalign box > <WRAP leftalign box >
-Both of the methods that we proposed are implemented using C++ for high efficiency. Although the systems are tested using standard image and question-answer datasetsthe evaluation can be extended to other datasetspotentially the ones from the IAB members. +We evaluated the proposed robust and efficient link prediction system with high level of performance in purpose of generating the possible hypothesis using biomedical literature. Howeverour model can be applied to a wide variety of problemswhich can be modeled as graphsFurthermorewe are working with Schumacher Clinical Partners to employ the link prediction model in predicting the medical conditions of patients.
- +
-We note that, the graph matching technique that we proposed in this project can also be adapted to other data types such as sound or videoAdditionally, the graph embedding into HST can also be applied on clustering tasks which might appear fields such as in online marketing or social networks.+
  
 </WRAP> </WRAP>
Line 36: Line 34:
 ===== Project - Documents ===== ===== Project - Documents =====
  
-{{filelist>*6a.001.tu*&sort=mtime&style=table&tableheader=1&showdate=1&showsize=1}}+{{filelist>*15.2*&sort=mtime&style=table&tableheader=1&showdate=1&showsize=1}}
  
 ~~DISCUSSION~~ ~~DISCUSSION~~
projects/year4/15.2.1566403855.txt.gz · Last modified: 2019/08/21 11:10 by sally.johnson