Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Technology Mapping with Large Language Models

Minh Hieu Nguyen, Hien Thu Pham, Hiep Minh Ha, Ngoc Quang Hung Le, Jun Jo

TL;DR

The paper addresses the challenge of mapping corporate technology portfolios from vast unstructured data, where keyword-based methods falter. It introduces STARS, a framework that fuses LLM-based entity extraction with Chain-of-Thought prompting and SBERT-based semantic ranking to identify and rank technologies for each company. Key contributions include a CoT-enhanced extraction process, a labeled technology dataset, and a SBERT-driven ranking mechanism that yields higher retrieval precision than baselines. Experiments on a Crunchbase-derived dataset demonstrate robust improvements in both company-to-technology and technology-to-company retrieval, highlighting STARS' applicability across industries and its potential impact on technology intelligence and strategic decision-making.

Abstract

In today's fast-evolving business landscape, having insight into the technology stacks that organizations use is crucial for forging partnerships, uncovering market openings, and informing strategic choices. However, conventional technology mapping, which typically hinges on keyword searches, struggles with the sheer scale and variety of data available, often failing to capture nascent technologies. To overcome these hurdles, we present STARS (Semantic Technology and Retrieval System), a novel framework that harnesses Large Language Models (LLMs) and Sentence-BERT to pinpoint relevant technologies within unstructured content, build comprehensive company profiles, and rank each firm's technologies according to their operational importance. By integrating entity extraction with Chain-of-Thought prompting and employing semantic ranking, STARS provides a precise method for mapping corporate technology portfolios. Experimental results show that STARS markedly boosts retrieval accuracy, offering a versatile and high-performance solution for cross-industry technology mapping.

Technology Mapping with Large Language Models

TL;DR

The paper addresses the challenge of mapping corporate technology portfolios from vast unstructured data, where keyword-based methods falter. It introduces STARS, a framework that fuses LLM-based entity extraction with Chain-of-Thought prompting and SBERT-based semantic ranking to identify and rank technologies for each company. Key contributions include a CoT-enhanced extraction process, a labeled technology dataset, and a SBERT-driven ranking mechanism that yields higher retrieval precision than baselines. Experiments on a Crunchbase-derived dataset demonstrate robust improvements in both company-to-technology and technology-to-company retrieval, highlighting STARS' applicability across industries and its potential impact on technology intelligence and strategic decision-making.

Abstract

In today's fast-evolving business landscape, having insight into the technology stacks that organizations use is crucial for forging partnerships, uncovering market openings, and informing strategic choices. However, conventional technology mapping, which typically hinges on keyword searches, struggles with the sheer scale and variety of data available, often failing to capture nascent technologies. To overcome these hurdles, we present STARS (Semantic Technology and Retrieval System), a novel framework that harnesses Large Language Models (LLMs) and Sentence-BERT to pinpoint relevant technologies within unstructured content, build comprehensive company profiles, and rank each firm's technologies according to their operational importance. By integrating entity extraction with Chain-of-Thought prompting and employing semantic ranking, STARS provides a precise method for mapping corporate technology portfolios. Experimental results show that STARS markedly boosts retrieval accuracy, offering a versatile and high-performance solution for cross-industry technology mapping.
Paper Structure (14 sections, 7 equations, 4 figures, 1 table)

This paper contains 14 sections, 7 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Works
  3. Preliminaries
  4. Problem Definition
  5. Mathematical Formulation
  6. Challenges in Technology Mapping
  7. Methodology
  8. Entity Extraction with Chain-of-Thought Prompting
  9. Technology-Company Retrieval via Semantic Ranking
  10. Experiments
  11. Datasets
  12. Metric: Precision at K (P@k)
  13. Evaluation
  14. Conclusion

Figures (4)

  • Figure 1: Overview of the Framework for Technology Landscape Mapping: STARS.
  • Figure 2: Overview of our Chain-of-Thought Prompts for Technology Extraction.
  • Figure 3: Effect of few-shot examples on P@k for company-to-technology retrieval.
  • Figure 4: Comparison of ranking methods for semantic matching. SBERT outperforms across all $k$ values.