Multi-modal transform-based fusion model for new product sales forecasting

Xiangzhen LI, Jiaxing SHEN, Dezhi WANG, Wu LU*, Yuanyi CHEN

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review


New product sales prediction is crucial for the digital economy as it enables businesses to make informed decisions about product development, inventory management, marketing strategies, and ultimately driving economic growth and innovation. In the digital economy era, traditional sales forecasting methods often struggle to address the unique challenges of forecasting demand for new products, primarily due to limited historical data and high levels of uncertainty. To address this challenge, we propose a multi-modal transform-based fusion model for new product sales prediction (M2TFM), which integrates multiple data sources (e.g., product images, attributes, text descriptions and context factors like holidays, weather and trends.) to predict new product sales with remarkable accuracy. The proposed method leverages diffusion embedding to fuse heterogeneous data modalities including images, text, and time series into a unified representation that models their complex interactions. By encoding multi modal data using Transformer self-attention, our approach is able to extract nuanced signals across modalities to make more accurate new product sales forecasts. We perform a comprehensive evaluation on a large e-commerce dataset with more than 10,000 fashion items, and the results demonstrate that the proposed method is more effective than existing state-of-the-art baselines for new product sales forecasting.
Original languageEnglish
Article number108606
JournalEngineering Applications of Artificial Intelligence
Issue numberPart F
Early online date23 May 2024
Publication statusE-pub ahead of print - 23 May 2024

Bibliographical note

Publisher Copyright:
© 2024 Elsevier Ltd


  • Attention mechanism
  • Diffusion modeling
  • Digital economy
  • Multi-modal fusion
  • New product sales forecasting
  • Temporal feature


Dive into the research topics of 'Multi-modal transform-based fusion model for new product sales forecasting'. Together they form a unique fingerprint.

Cite this