Tiran Software - PowerPoint PPT Presentation

About This Presentation
Title:

Tiran Software

Description:

Features of verbs, adjectives, nouns... Current Status (cont. ... SVM-Light. WordNet. JWNL. TDK / Zargan. Zemberek, PostgreSQL. Tools & Resources. Any Questions? ... – PowerPoint PPT presentation

Number of Views:30
Avg rating:3.0/5.0
Slides: 16
Provided by: rx64
Category:

less

Transcript and Presenter's Notes

Title: Tiran Software


1
Tiran Software
  • -TURKUAZ Project-
  • RadeX
  • Tahir Bilal
  • Onur Deniz
  • Soner Kara
  • M. Mert Karadagli

Assistant Umut Erogul Instructor
Meltem T. Yöndem
2
Outline
  • Problem Definition
  • Important Aspects
  • Our Approach
  • General Structure
  • Analyzer Component
  • Searcher Component
  • Current Status
  • Prototype
  • Tool and Resources
  • Q/A

3
Problem Definition
  • Billions of radiology reports
  • Unfortunately, they are stored in free-text
    format
  • Hard to search and retrieve
  • Need for searchable information

4
Important Aspects
  • Text Mining
  • NLP
  • Information Extraction
  • Morphological Analysis
  • Named Entity Recognition
  • Machine Learning
  • Neural Networks, Decision Trees ...

5
Our Approach
  • RadeX, Radiology Data Extractor will enable..
  • Modular machine learning component
  • Support for internal/external dictionary
    connection
  • Template-based approach for finalizing

6
General Structure
7
General Structure (cont.)
  • Analyzer Component
  • Preprocess free text
  • Look-up internal and external lexicons
  • Gives semantic to words
  • Extracts searchable data
  • Searcher Component
  • Send query strings to database
  • Retrieve corresponding information

8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
Current Status
  • Preprocessing.
  • Connecting and using external sources.
  • Database implementation.
  • Applying SVM to unrelated but tagged corpus.

12
Current Status (cont.)
  • Mapping Turkish terms to English translations.
  • Finding stem of unknown words.
  • Constructing lexicons.
  • Features of verbs, adjectives, nouns...

13
In Prototype we will be able to...
  • ..decompose reports into sub-parts, sentences and
    words,
  • .. analyze words using Zemberek and a stemmer.
  • .. give semantics to words via internal/external
    lexicons
  • .. extract simple information using pre-defined
    templates

14
Tools Resources
  • SVM-Light
  • WordNet
  • JWNL
  • TDK / Zargan
  • Zemberek,
  • PostgreSQL

15
Any Questions?
Write a Comment
User Comments (0)
About PowerShow.com