Top Features of Recommender System
Data Ingestion, simplified
SOLUS AI takes in online and offline transactions, online behaviors, preferences, location and product data, and more. These multiple signals help it come up with personalized recommendations that are relevant in all contexts, for each customer.
Learning Algorithms
At the heart of the Recommender system are multiple recommender algorithms, each of which caters to a different strategy. This approach, combined with a “Recommender of Recommenders”, ensures that we solve for different contexts ranging from cold-start to the data rich.
For instance, Genome Matching is a strategy that looks at “what do people like you buy”, while a fairly distinct strategy is “what are your past preferences”. Both of these (and more) co-exist within SOLUS.
Outputs as Stories
Our recommender output are made available as “stories” – marketing-friendly ways to use the outputs from the machine learning system, without haveing to get into what the underlying algorithm is. want to cross-sell? want to drive exploration? want to push reorder? There’s a story for every use case with solus.
- Top N Recos
- Cross Sell
- Reorder
- In Cart Based
- Wishlist Based
- Fav. Cat/ Brand
- Exploration
- Smartbasket
- Recos Within A Category
- People Like You
- Trending In City/ Season
Recommendations Deployment
We’ve made deploying recommendations simple, and super-quick. SOLUS has business layers where a user can include or exclude recommendations, add stock and inventory checks or even bump up promoted products. Finally at the deployment end we have multiple options ranging from our custom short-link + landing page deployment to Recos as a Service.
Our Recommenders At Work
A Few Of Our Clients
E-301, 3rd Floor, Lotus Corporate Park, Off Western Express Highway, Jay Coach Area, Laxmi Nagar, Goregaon (E), Mumbai – 400063.
slath@solus.ai