Build workflows in KNIME (text mining)

Closed Posted Jan 4, 2014 Paid on delivery
Closed Paid on delivery

Build workflows in KNIME

ONLY MAKE AN OFFER WHEN YOU HAVE ALREDY DONE KNIME DEVELOPMENT PROJECTS!!

ALL OTHER OFFER WILL BE IGNORED.

1. Extract content

==================

Extract content from PDF documents in a folder

2. Isolate textblocks

=====================

- All textblocks in the document are identified (isolated)

- The start- and end location of each textblock in the document is identified

- A textblock can contain several paragraphs, sentences, words or a single word.

3. Label textblocks

================

- Automatically assign label to each text block, based on specific keywords for that each label.

- The specific keywords are retrieved for a table in a mySQL database. These keywords are used to see if they match the words in the textblock. If a textblock contains less then 4 words, then there must be a 100% match. If the textblock contains 4 words or more, the match van be partial. A treshold value (x%) is used. So for example if more then 85% percent of the textblock matches the keywords of a label in the mySQL table, there is a succesful match and this specific label is used for this textblock.

4. Store textblocks

===============

Each textblock is stored in a table in a mySQL database with these values:

- documentname

- label

- textblock

- startposition

- endposition

Data Mining

Project ID: #5283994

About the project

1 proposal Remote project Active Feb 10, 2014

1 freelancer is bidding on average $333 for this job

invictos

Sir i have done work in KNIME and have experience in data mining. Looking forward to work with you. Regards Mujtaba Aslam

$333 USD in 3 days
(0 Reviews)
0.0