Big Data and Machine Learning at Kohl's Department Stores

Original

The landscape of retail is changing; companies are using more and more data-driven decisions. Machine learning and decision sciences allow companies to understand their customers and stores better, and enable them to learn how to make more informed decisions around various landscapes of retail, including merchandising, marketing, fulfilment.

Abhik Banerjee, Staff Data Scientist at Kohl's Department Stores, will be presenting at the Machine Intelligence Summit 23-24 March in San Francisco. His session will cover how Kohl's use big data and machine learning technologies, including the algorithms used and deployed in production related to recommendations and customer segmentations, by using unsupervised learning approaches, and other supervised learning problems on scale using Spark and Hadoop. 

We interview Abhik ahead of the Summit to learn more about his role and his thoughts on machine learning. 

 - Tell us more about your work at Kohl's

At Kohl's the company is moving in a great direction, where every function or every area in the retail business wants to be data driven, and use data not only to make predictive but provide prescriptive intelligence to the data. Some of the areas we are powering at Kohl's, apart from Recommendations, is to understand our customers and stores better. Let me go a bit deeper into these areas, 2 of the main assets for Kohl's are our customers and to make our stores profitable. We want to personalize the whole experience for our customers, throughout their whole journey, from when they land on our Kohl's Homepage, where there will be personalised deals, personalised products recommendations on every page, and personalized offers when our ML pipeline models predict that the customer is going to abandon our carts. We build our whole marketing campaigns by understanding more about our customers, and their whole personalized affinities towards particular brands, sizes, colors, categories, etc.

We want to use Data Science to decide on the whole merchandising allocation, which not only takes our internal data sets like transaction, browse - clickstream behavior, but also external data sets like weather, macro-economic conditions, social data, etc.

The other vision is around Smart Store. Kohl's has close to around 1100 stores and we want to use all the data available in and around the store to make our stores smarter, Our CTO Mr. Ratnakar Lavu, recently highlighted this vision in the Time Magazine, which really lays down the details of this initiative. Imagine a store when a customer walks into the store, the whole store is personalized as per their needs; the store has the right inventory and the right products to cater to the group who are visiting the stores. We understand our customers around the store and their behaviour so well that we can personalize their whole experience and give the best possible experience, in the form of an Omnichannel experience.

We want to become the best technology company in the domain of departmental store retail, we also have a few initiatives going on in the Deep Learning area, and power experience using image-based recommendations, as well as a lot of text mining and other use cases which heavily employ Deep Learning to the core.

- What do you feel are the leading factors enabling recent advancements in ML?

Some of the leading reasons for so much development in the areas for advancement of ML is due to the recent availability of large amount of data sets, and also the high powered computing platforms like GPU. It's comparatively much easier now to run deep neural network models with the current infrastructure as it was 8 years ago. Also some of the recent contributions on open source communities, like tensorflow, caffe, etc, it has really become very easy for building complex models on top of this framework, and improves a lot of the advancement in the ML domain.

- What present or potential future applications of Machine Intelligence excite you most?

I am really excited about artificial intelligent bots, and they are really near and dear to my heart. With the advent of machine learning and artificial intelligence being a part of most of our business, I feel retail is one of the key areas where the whole machine learning and artificial intelligence has a huge scope of improving the way we see things. Data is the new oil and retail has huge amounts of data which to some extent had been used for making informed decisions, but I still feel it has not realized it true potentials yet, especially when one has so much data about customers being gathered by IoT devices, their social behaviours and interactions, and their various other touch points in their apps and mobile devices. Imagine a world where you start having your personal assistants which not only helps in your daily activities but which guides you for your next fashion-based recommendations, starts learning your preferences and helps you to plan your trips, what to wear to your trips, what to carry, which store to go to for the best deals, and something which you have a strong predilection too, helps with gift based recommendations for your family and relatives, and make this a whole connected personalized experience and makes it better as it learns more and more about you and people around you.

I don’t think we are that integrated at present, but we are getting there and I am very optimistic, we will reach it one day.

- Which industries will be most disrupted by Machine Intelligence?

I feel almost all the industries will be disrupted by machine learning and Machine Intelligence in the forthcoming future. Something which Dr. Andrew Ng recently mentioned in one of his talks really resonate with me very well, where he mentioned AI and Machine Intelligence to be similar to how electricity was, and the way it transformed many industries. As we make the technology of Machine Intelligence better, we should see areas such as Retail, Telecom, Marketing, Customer Support systems, and almost every aspect being disrupted using Machine Intelligence.

- What developments can we expect to see in Machine Intelligence in the next 5 years?

I am very optimistic to see a lot of growth in Machine Intelligence and machine learning in the next 5 years, as per a recent post, for tourist destinations, people don't need IDs but use face detection to get entry. There are numerous examples of how things are already being transformed using AI and Machine Intelligence. I feel as the research in these areas make the technology better in terms of Machine Intelligence, various areas and our daily lives will be greatly impacted by these changes.

Discover advances in machine learning and artificial intelligence from the world's leading innovators at the Machine Intelligence Summit, 23-24 March in San Francisco, which will be running alongside the Machine Intelligence in Autonomous Vehicles Summit. 

Learn from the industry experts in speech & image recognition, natural language processing and computer vision. Explore how AI will impact transport, manufacturing, healthcare, retail and more. Speakers include Amy Gershkoff, Chief Data Officer at Ancestry; Erik Schmidt, Senior Scientist at Pandora; Stacey Svetlichnaya, Software Development Engineer at Flickr; Rudina Seseri, Founder and Managing Partner at Glasswing Ventures and Andrew Owens, PhD Student at MIT. 

Early Bird Pass expires on 27 January, don't miss out on this big saving. 

For more information and to register, visit the event website here.

Original

Big Data Machine Learning Data Mining IoT AI Machine Intelligence Summit


0 Comments

    As Featured In

    Original
    Original
    Original
    Original
    Original
    Original

    Partners & Attendees

    Intel.001
    Nvidia.001
    Ibm watson health 3.001
    Acc1.001
    Rbc research.001
    Twentybn.001
    Mit tech review.001
    Kd nuggets.001
    Facebook.001
    Maluuba 2017.001
    Graphcoreai.001
    Forbes.001
    This website uses cookies to ensure you get the best experience. Learn more