Home
SWIS-it, a semantic search tool, is a crawler-based indexing and retrieval system for web documents. SWIS-it discovers, digests, analyses and indexes the documents to the most appropriate concepts and domain.
To create a platform which retrieves and categorises documents into semantic categories and store index of these documents.
Search engines are programmed to rank websites based on some combination of their popularity and relevancy in the information they provide and the underlying assumptions about the technology.Many search engines provide customised results based on user's activity history. As a result, websites tend to show only information that agrees with the user's past viewpoint. It also brings out a lot more results than required by the user.
SWIS-it, a search engine is one such solution for bringing out results to improve search accuracy by understanding the searcher intent and the contextual meaning of terms as they appear in the searchable dataspace to generate more relevant results.
- Extend Google search capabilities to semantic search for documents.
- Help users provide the most appropriate results.
- Minimise the time to find relevant results.
SWIS-it works on the assumption that there are certain semantic categories of documents that the user is interested in. Some of these are:
- Introductory material
- Solved examples
- Theoretical reference
- Advanced implementation
Each semantic category can be described by a graph of:
- Terms – words or phrases encountered in documents of this type
- Structures – document structures
The graph links category to terms. There are two kinds of links – indicators and counter-indicators. Each link has a property weight.
A user sets up a search query by defining:
- Concepts to search for
- Semantic categories
- URLs to search in
- Google app credential to use
User can then search the index for documents based on queries that combine concepts and categories
SWIS-it is built on top of Java with Neo4J as the graph database and Angular2 as front-end.
Presentation: The front-end of our application is a Single Page Application (SPA) developed using Angular2. As we are using Angular2, we have a highly modular component-based architecture. First-time users of the application will instantly see a server rendered view, which greatly improves perceived performance, speed and the overall user experience.
Middleware: We used Spring framework.
We Used:
- MongoDB to store user information as JSON documents.
- Neo4J to store our graph ontology divided into Concept and Intent graph.
- Redis to store the concepts and intents from Neo4J when the application starts.
Abbreviation for SWIS-it
- Semantic Web Intent Search-it