Visualizing the Performance of Computational Linguistics Algorithms

Author

Eick, Stephen G. ; Mauger, Justin ; Ratner, Alan

Author_Institution

SSS Res., Inc., Naperville, IL

fYear

2006

fDate

Oct. 31 2006-Nov. 2 2006

Firstpage

151

Lastpage

157

Abstract

We have built a visualization system and analysis portal for evaluating the performance of computational linguistics algorithms. Our system focuses on algorithms that classify and cluster documents by assigning weights to words and scoring each document against high dimensional reference concept vectors. The visualization and algorithm analysis techniques include confusion matrices, ROC curves, document visualizations showing word importance, and interactive reports. One of the unique aspects of our system is that the visualizations are thin-client Web-based components built using SVG visualization components

Keywords

computational linguistics; document handling; portals; program visualisation; software performance evaluation; AJAX; ROC curves; SVG visualization components; analysis portal; computational linguistics algorithm performance visualization; confusion matrices; document classification; document clustering; document visualizations; interactive reports; scalable vector graphics; thin-client Web-based components; Algorithm design and analysis; Clustering algorithms; Computational linguistics; Content based retrieval; Information retrieval; Performance analysis; Portals; Storage automation; Text analysis; Visualization; AJAX; ROC curves; SVG; confusion matrices; document categorization; thin-client;

fLanguage

English

Publisher

ieee

Conference_Titel

Visual Analytics Science And Technology, 2006 IEEE Symposium On

Conference_Location

Baltimore, MD

Print_ISBN

1-4244-0591-2

Electronic_ISBN

1-4244-0592-0

Type

conf

DOI

10.1109/VAST.2006.261417

Filename

4035760