With more than 70 different services in the Rakuten Group globally, our data science teams cover diverse initiatives in both the online and offline worlds. Offline Geodata and Geo Science applications are key to optimization and overall success in the fields of Logistics, Site Planning, Area Marketing and improving Customer Experience and many more. In this talk, we will cover some of the Geo initiatives at Rakuten and discuss some interesting questions around the viability for online data to fully represent a user and how to potentially rectify that via offline data and data science.
Customer & Product Sciences
Product Matching In E-Commerce, Product Matching is one of the fundamental problems for various use cases like (1) Competitive pricing of products, (2) Deduplication of products in the catalog, (3) Grouping items from various merchants (4) Recommending products. The requirement is to match a product accurately against a catalog spread across tens of thousands of taxonomy nodes and millions of items. Product matching results must be accurate, and the margin for error is minimal to use Product Matching across use cases. In this talk, we touch upon the “Six honest serving-men” by Rudyard Kipling –What, Why, How, Where, When, and Who of Product Matching. We share the overall approach, experience in designing and orchestrating Deep Learning models with a scalable architecture to achieve the required results. Here we have approached the problem at the grass-root level consisting of five stages (1) Identifying attributes per taxonomy node (2) Classification of Products (3) Attribute Enrichment from NER (Text) and Image feature extraction (4) Search against multiple indices and filter results for mandatory attributes (5) Re-rank to improve the relevancy of shortlisted results. We also address Product data quality, measured at every stage to improve the overall performance of Product Matching, towards meeting business needs.
The recent advances in artificial intelligence are mainly attributed to deep learning models, which are intrinsically data-hungry and compute-hungry. Moreover, such deep learning models also suffer from black-box-ness, vulnerable understanding and reasoning ability, and intrinsic bias. All these factors limit its domain application. Explicit knowledge represented in knowledge graphs has a lot to offer. It could potentially drive the next surge in AI by addressing all of the above concerns. Let us brainstorm “Role of Knowledge Graphs in AI” with the education domain case study researched and developed at Embibe.
Learn how retailers use AI-based product enrichment techniques and image recognition models to make products more searchable. Understand how using AI smart search and complementary product recommendations like “shop the look” can help eCommerce companies engage consumers and drive increased revenue.
The ever-evolving economic scenarios, regulatory framework, changing consumer behaviour, data availability are additional factors besides the analytical consideration for a credit score. Therefore, a credit score using the robust analytical technique while addressing these factors is key for its adoption.
Rakuten group operates more than 70 different services, such as e-commerce, fintech and mobile services, our data capture customer behaviours with various aspects. By fully utilising Rakuten data, we’re building customer models for understanding customers. In this talk, we will introduce several research projects such as (1) customer attribute prediction, (2) customer relationship identification, (3) measuring customer credibility and (4) prospective user extraction.
Learn how the world’s second-largest home improvement retailer helps customers love where they live. This talk will be focused on a couple of case studies where Lowe’s has been able to improve customer engagement and, thereby, revenue through the use of ML and personalisation in the context of offline retail. This will also delve into challenges faced and the key learnings.
Unravel Data is an AI-enabled product to simplify the observability of modern data platforms and is used today by the largest brand names in Fortune 500. This talk covers the hard lessons I learned over the years in developing dozens of Data+AI products from idea to production — things that can go wrong across ML problem definition, dataset selection, data preparation, model design, training, and operationalization in production. I wrap up the talk with key patterns you can follow to build your AI/ML products successfully.
