I have checked Lucene's way of indexing files. But I would need to index key value pairs. Like I have many (word,sentence) combinations in a single file or few number of files. Most of the cases, it is in the single file.
I would like to search the sentence based on the word which I use for searching. What's the mechanism that I need to use with Java?
Something similar to Sequence file in Hadoop. Without parsing through entire data, we should be able to get the value for every key available.