Graph Analytics & Machine Learning for AI
Crawl and index any kind of content in a relationship-based approach. Do very fast, in-depth graph analysis of the same. For ex. give me city-wise breakup of all relatives of my friends. This query will typically take multiple (minimum -> 1 for friends + 1 for relatives + numUniqueCities for city names) queries in any datastore and will be complex for a developer to write and test. It will naturally be heavy on the network, CPU, and RAM of the system. Using EG this can be executed in a single hit, thanks to index time joins.
One of our clients was into content intelligence domain. They would crawl popular URLs shared, liked, commented on over 40 social networks and some news websites all throughout the day, tracking literally every retweet or share. Next iteration would be to extract metadata like author, date, keywords, domain, industry, sentiment, abstract, etc from those URLs and index the same. The third step would be to extract more information about the author - his Linkedin profile, Twitter profile, etc. The third step would be to classify the author as a sharer, broadcaster, retweeter, original poster, influencer, etc. Another thing would be to identify all popular keywords or entities (through semantic extraction) being talked about in any domain. This would tell the marketeers of any domain what is trending in that domain today, this week, this month, etc. so that they can talk about that. They can also identify the community or within any domain and target them to be their brand ambassadors. They could identify top content creating websites and blogs and ask them to write about them, or put an ad there.
There is SO MUCH that you can do when you have content. All you need is creative thinking and a platform to make that thinking into reality, with good speed and cost-effectiveness. This is what we help you do with Elasticgraph.