Resurgence of Retailers in a Post-Covid 19 World

Inspiration Difficulty faced by physical retailers in a global pandemic . Many of them were forced to close down due to lack of customers and reduced demand of goods . The idea of our proposal is to basically encourage physical retailers to adopt e-commerce supported by Artificial Intelligence . Although the initial cost of adopting into a new business model may be costly , more sales can be generated in the long run as well as access to a wider customer base via the use of social media .

What We Learned We found out that recommender systems are being used by tech giants like Netflix, Amazon and many more to target their content to a specific audience . Further research reveals that an algorithm called collaborative filtering is incorporated into many recommender systems . Collaborative filtering algorithm is a method of making automatic predictions (filtering) about the interests of a user by collecting preferences or taste information from many users (collaborating). It analyses how similar the tastes of one user is to another and makes recommendations on the basis of that.For instance, if user A likes movies 1, 2, 3 and user B likes movies 2,3,4, then they have similar interests and A should like movie 4 and B should like movie 1. This makes it one of the most commonly used algorithm as it is not dependent on any additional information.

How we built our project We decided to incorporate the above algorithm in the context of e-commerce dataset by using the movie recommender system as reference . The algorithm should make item recommendations to users based on the preferences of many other users .

Challenges Faced -Access to suitable e-commerce dataset -Many commercial recommender systems are based on large datasets. As a result, the user-item matrix used for collaborative filtering could be extremely large and sparse, which brings about the challenges in the performances of the recommendation.The data sparsity can cause the cold start problem. As collaborative filtering methods recommend items based on users' past preferences, new users will need to rate sufficient number of items to enable the system to capture their preferences accurately and thus provides reliable recommendations.

Built With

• fastai
• machine-learning
• python