Information retrieval system.

1. Work on a local document corpus.
2. Parse and index documents using inverted file format
3. Make use of stemming and stop lists.
4. Implement matching based on tf-idf algorithm.
5. Results are displayed based on relevance ranking.

• src/Source.cpp Program source codes
• src/IndexCreator.ext Compiled, exectuable codes. Generate from source.cpp
• data/stop.txt Stop words get from internet.
• data/TD_IDF_INDEX.txt Web stite conetent get from Steves Institue site by web scrawler.
• data/test/... Different training results based on size.

存数据 已知 term， doc， tf， idf 建一个hash mytable( key=term value=doc) For 循环读取每行信息 If (tf!=0) Mytable[key]=vector.push(doc) End for 返回mytable

查询 term 返回 mytable[term];

查询 term and term1 或者 term term1 返回 mytable[term] && mytable[term1]

查询 term or term1 返回 mytable[term] || mytable[term1]

查询 not term 返回 whole doc list - mytable[term]

用栈处理查询语句优先级 将处理后的查询语句循环查询