We were reviewing how MongoDB enables integration with social engineering applications while also enabling data and service integration for the user data inflow with APIs for data and services during peak annual sales season. At the heart of their store, they can categorize data based on merchandising, content, inventory, customer, channel, sales and fulfilment, insight and social. Out of these merchanding, inventory, customer and insight are the most heavily used for peak holiday sales season.  In addition,  supply chain management system and data warehouse also integrate well with this Storage.

The information layer improves experience with look and feel, Navigation, customization, personalization, branding, promotion, chats and ads. Each of these is a separate initiative.  The customer's perspective helps us understand their context. For example, a user would like to browse and search the store online. This is part of his research. The shopping cart helps him select the items. The checkout experience is streamlined towards purchase.  The tracking of the item improves the receiving experience.  There is feedback loop and maintenance considerations with use of item.  The dialog with the customer is the focus of the assistance efforts. All these customer experiences translate to sofware interactions in two or three major categories on the enterprise side. For example,  the Market/Offer will require services separate from sell/fulfill.  The sell/fulfill will required data servces different from insight. Note that the services are easily enumerable within each category to be different from each other while they share the same information management backend which is facilitated with MongoDB which provides a one-stop shop for information management specifically in the areas of merchandising, inventory, customer and insight  The merchandising fulfils market/offer services such as guide, offer, semantic search, recommend, rule-based decisions, pricing and coupons. The Sell/fulfill provides ways for order, payments, fraud detection, fulfilment and business rules. The Insight provides ways for session capture and activity monitoring.

#codingexercise continued

Yesterday we described finding the largest submatrix in a binary matrix. 

We said the conventional way was to find connected components. Today we explain this:

int maxsize = 0
for (int i = 0; i < ROW; i++)
   for (int j = 0; j < COL; j++)
      if ( A[i,j] == 1 && !visited[i, j])
   {
        int size = 0;
        count++;
        DFS(A, ROW, COL, i, j, ref visited, ref size);
        if (size > maxSize)
             maxSize = size;

   }
This can be modified to record the cells of the submatrix contributing to the maximum size.
Given a cell in a connected component, we can find the top-left corner of the bounding submatrix with the minimum values for the x and y co-ordinates.

The maximum values of the x and y co-ordinates leads us to the bottom-right corner of the bounding submatrix