To read the original post please visit this link on dev.to
Almost 2 years I came across Elasticsearch and was astonished after having a look at the complex functionalities that it provide just out of the box. And since then I have digging into book, article, company tech blogs to find out how such Search/IR systems works.
While going through many articles I came across "Reverse Indexes" and my love for it grew every day since then.
The amount of resources left over the internet to read about Inverted Index and Search System are very limited. So I thought to give my 5 cents back to our tech community via these articles explaining the core concepts of Reverse Indexes and its implementation.
When we understand the underlying technology to the core, then only we can really appreciate the "out of the box" functionalities that we get in our frameworks/DBs. ✌️
So we will be covering following topics in this series one in each post, with code examples whenever required:
- Introduction to Inverted Indexes
- Inverted index Component 1 - Dictionaries
- Inverted index component 2 - Posting Lists
- Tokenization
- Structure of Inverted index
- Creation of the Reverse index
- Storing index-Data over disk
- Retrieval and Ranking - Vector Space Model
- Inverted Index Compression
- Distributed Index Creation
- Dynamic Index Creation
- And More...
I will be writing these articles one by one in the coming month. before 6th July 2020. And will be attaching links to the articles list mentioned above.
You reviews/feedbacks/important are most welcome.
TODO code improvs: Use Java Modules instead of a single src folder. I am new to java, will be having different iteration for code implementations to make it better.
Stay Tuned. Stay Safe.