Prediction of future customer needs using machine learning across multiple product categories.

ABSTRACT: In recent years, computational approaches for extracting customer needs from user generated content have been proposed. However, there is a lack of studies that focus on extracting unmet needs for future popular products. Therefore, this study presents a supervised keyphrase classification model which predicts needs that will become popular in real products in the marketplace. To do this, we utilize Trending Customer Needs (TCN)-a monthly dataset of trending keyphrase customer needs occurring in new products during 2011-2021 across multiple categories of Consumer Packaged Goods e.g. toothpaste, eyeliner, beer, etc. We are the first study to use this specific dataset and employ it by training a time series algorithm to learn the relationship between features we generate for each candidate k

SUBMITTER: Kilroy D 

PROVIDER: S-EPMC11346667 | biostudies-literature | 2024

REPOSITORIES: biostudies-literature