-
Notifications
You must be signed in to change notification settings - Fork 0
/
Readme.txt
19 lines (14 loc) · 958 Bytes
/
Readme.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Java search engine. Preprocessing of the corpus and searching by query.
Koren Ishlach
Tal Frimerman
*******************************************
************************************
*************************
***************
**Few notes for the second part**
1. The output path needs to contain the output path used for pre-proccessing the corpus (or wherever the output files are stored).
2. If the ipnut path will not contain a directory with "5 stop_words" file, the stop words that will be taken into account are the words from the original file.
3. When running with semantically similar words: it is *not necessary* to have a working internet connection, but it is recommended for better results (The program
will try to use a semantically similar word search that requires internet connection, and if that fails it will use a method which does not require internet connection ,
which from our tests gives less good results).
Enjoy!