DSearch™ is a full solution for creating vertical search engines in specific domains. You just define the domain to us verbally and we will create the necessary ontology for crawling and extraction.
The solution has the following main components:
- Domain specific crawler that can crawl websites and determine their relevancy to the search. Each domain name will be given an estimate degree or relevancy to our domain. Crawler will store all of the information in an easy-to-retrieve database.
- Extraction Phase where the engine will be extracting features for the products. Those features will be grouped depending on their specific values or depending on the semantic of the field.
- Ontology creation: where keywords will be grouped into an ontology tree.
- Identification: Extracted fields will be grouped automatically and uniquely identified into appropriate fields.
- Indexing: standard stemming, tokenization and indexing procedures.
DSearch™ is a vertical search engine that is domain specific. The search engine will be available immediately. DSearch™ engine will continue on crawling and updating its database in real-time (almost one update per hour) and data will be updated in the results as they are crawled, extracted and identified.
The advantages of DSearch™ are:
- Easy to configure and run.
- Solid solution and results are 100% correct even for new domains.
- Low on resources consumption. You don’t need a full cloud of servers. If the number of pages to be indexed is less than 10 million pages then couple of server will be sufficient.
- Produce a full facet search on the data crawled.