Dynamic technology impact analysis: A multi-task learning approach to patent citation prediction

Youngjin Seol,Jaewoong Choi,Seunghyun Lee,Janghyeok Yoon

2024-11-14

Abstract:Machine learning (ML) models are valuable tools for analyzing the impact of technology using patent citation information. However, existing ML-based methods often struggle to account for the dynamic nature of the technology impact over time and the interdependencies of these impacts across different periods. This study proposes a multi-task learning (MTL) approach to enhance the prediction of technology impact across various time frames by leveraging knowledge sharing and simultaneously monitoring the evolution of technology impact. First, we quantify the technology impacts and identify patterns through citation analysis over distinct time periods. Next, we develop MTL models to predict citation counts using multiple patent indicators over time. Finally, we examine the changes in key input indicators and their patterns over different periods using the SHapley Additive exPlanation method. We also offer guidelines for validating and interpreting the results by employing statistical methods and natural language processing techniques. A case study on battery technologies demonstrates that our approach not only deepens the understanding of technology impact, but also improves prediction accuracy, yielding valuable insights for both academia and industry.

Machine Learning,Artificial Intelligence

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is the deficiency of existing technology impact analysis methods in dealing with the temporal dynamics of technological impacts and the interdependencies between different time stages. Specifically, existing machine - learning (ML) methods are often difficult to capture the changes of technological impacts over time and the interactions of these impacts between different periods. To solve these problems, the paper proposes a multi - task learning (MTL) approach to improve the prediction performance of technological impacts within different time frames by sharing knowledge and simultaneously track the evolution process of technological impacts. The core contributions of the paper are as follows: 1. **Quantitative Definition and Pattern Recognition**: Conduct citation analysis at different time stages to quantify technological impacts and identify patterns. 2. **Design and Development of MTL Model**: Use data of multiple patent indicators changing over time to design and develop MTL models to predict the number of citations. 3. **Explanation of Changes in Key Input Indicators and Patterns**: Use the SHapley Additive exPlanation (SHAP) method to explain the changes in key input indicators and their patterns at different time stages. Through these methods, the paper not only deepens the understanding of the dynamics of technological impacts but also improves the accuracy of prediction, providing valuable insights for academia and industry. For example, in the case study of battery technology, this method can not only predict the impact of technology more accurately but also reveal which features are most important for long - term and short - term technological impacts.

Dynamic technology impact analysis: A multi-task learning approach to patent citation prediction

Modeling Scholarly Collaboration and Temporal Dynamics in Citation Networks for Impact Prediction

Design of reliable technology valuation model with calibrated machine learning of patent indicators

Predictive modeling for technology convergence: A patent data-driven approach through technology topic networks

Assessing future technological impacts of patents based on the classification algorithms in machine learning: The case of electric vehicle domain

Early screening of potential breakthrough technologies with enhanced interpretability: A patent-specific hierarchical attention network model

On Predictive Patent Valuation: Forecasting Patent Citations and Their Types.

Patent Citation Dynamics Modeling via Multi-Attention Recurrent Networks

PTNS: patent citation trajectory prediction based on temporal network snapshots

Artificial Intelligence Technology analysis using Artificial Intelligence patent through Deep Learning model and vector space model

A Text Mining-Based Approach for Modelling Technical Knowledge Evolution in Patents

Identification of Technology Transfer Potential Based on Patent Dynamic Characteristics

Prediction of Emerging Technologies Based on Analysis of the U.S. Patent Citation Network

Identifying dynamic knowledge flow patterns of business method patents with a hidden Markov model

Learning Both Dynamic-Shared and Dynamic-Specific Patterns for Chaotic Time-Series Prediction

Technical Progress Analysis Using a Dynamic Topic Model for Technical Terms to Revise Patent Classification Codes

Development of Patent Technology Prediction Model Based on Machine Learning

Introducing DynaPTI–constructing a dynamic patent technology indicator using text mining and machine learning

A Study on Technological Dynamics in Structural Health Monitoring Using Intelligent Fault Diagnosis Techniques: A Patent-Based Approach

Modeling technological topic changes in patent claims

AI for Patents: A Novel Yet Effective and Efficient Framework for Patent Analysis