The retail business is getting back on track and has been witnessing steady growth after the dismal impact of the third wave. There has been buoyancy in the market with the removal of lockdown restrictions. After a long time of distress and uncertainty, things are getting back to normalcy as businesses have started taking pertinent steps to resume operations and focus on sales, marketing, and inventory management. The realization of digital transformation coupled with the indispensable role of artificial intelligence (AI) has been one of the major outcomes of Covid-19 implications on the retail sector and the vast possibilities and opportunities it can create with such transformations. With the emergence of e-commerce, buyers experienced the first crucial shift that successfully made it possible for them to buy things from anywhere at any time.
Successful conversational search systems can present natural, adaptive and interactive shopping experience for online shopping customers. However, building such systems from scratch faces real word challenges from both imperfect product schema/knowledge and lack of training dialog data.In this work we first propose ConvSearch, an end-to-end conversational search system that deeply combines the dialog system with search. It leverages the text profile to retrieve products, which is more robust against imperfect product schema/knowledge compared with using product attributes alone. We then address the lack of data challenges by proposing an utterance transfer approach that generates dialogue utterances by using existing dialog from other domains, and leveraging the search behavior data from e-commerce retailer. With utterance transfer, we introduce a new conversational search dataset for online shopping. Experiments show that our utterance transfer method can significantly improve the availability of training dialogue data without crowd-sourcing, and the conversational search system significantly outperformed the best tested baseline.