Skip to main content

HMS Case Study

#industry #finance #lifesciences #other

Identify Pharmaceutical Trends Earlier with AI-Powered News Analysis

Continuously analyze pharmaceutical news, automatically identify new topics, and track trends over time.

Management Summary

An international pharmaceutical company wanted to systematically analyze a growing volume of pharmaceutical news sources. The project demonstrates how AI can be used in the pharmaceutical industry for continuous news and trend analysis. The platform developed by HMS analyzes unstructured news data, automatically clusters topics, and visualizes their development over extended periods. This provides business units with a more reliable foundation for market monitoring, reviews, and management updates.

Project summary

#lifesciences
Internationaler Pharmakonzern

Internationaler Pharmakonzern

Branche: Pharma
Projektstart: 06/2025
Schwerpunkt: Data Science & Machine Learning
Use Case: Natural Language Processing

Projektziel

Aufbau einer Plattform zur kontinuierlichen Identifikation, Bewertung und Beobachtung relevanter Themen und Trends aus heterogenen Newsquellen.

Wichtigste Kernfunktionen

  • Kontinuierliche Analyse pharmazeutischer Newsdaten
  • Hypothesenfreie Themenidentifikation mit Machine Learning, NLP und generativer KI
  • Relevanzbewertung und Visualisierung thematischer Entwicklungen über Zeit
  • Nachvollziehbare Exploration von Themen, Quellen und einzelnen Artikeln

Tech Stack

python machine learning generative ai openai models & apis react amazon dynamodb aws lambda aws step functions

It is not enough to identify relevant topics early on. What matters most is that subject matter experts can understand the development, sources, and relevance.

Christoph Bergen

CoE Lead for GenAI at HMS

The Starting Point

Employees at pharmaceutical companies monitor a large volume of news, market information, and industry developments on a daily basis. As the amount of available information increases, so does the effort required to fully grasp relevant topics and reliably assess their significance.

Manual research often results in gaps because not all sources can be systematically evaluated. Assessments also depend on which topics are already known and what is being actively searched for. Even AI-powered chat solutions are primarily designed to address specific questions. As a result, new or unexpected developments are more likely to go unnoticed.

Simple process adjustments were therefore insufficient. Identifying new topics and long-term trends required continuous, scalable analysis of large volumes of unstructured content.

Zwei Projektmitarbeitende besprechen auf einem Laptop die grafische Auswertung der KI-gestützten Pharma-Newsanalyse.

The HMS Solution

AI-Powered Data Analysis for Pharmaceutical News and Trends

HMS developed a platform for the automated analysis and structuring of pharmaceutical news data. The solution continuously processes new content, identifies relevant topics, and provides a transparent, traceable overview of their development.

Data Integration and Processing

The platform integrates unstructured text data from various news sources into a unified data model. New content is standardized, automatically processed, and made available for analytical purposes.

The backend architecture is based on AWS. It uses relational and NoSQL databases as well as orchestrated microservices for data ingestion, processing, and delivery.

Hypothesis-Free Topic Identification

The platform does not require any predefined search terms or topics. It analyzes the news database and automatically groups articles into thematic clusters.

The combination of machine learning, NLP, and generative AI supports the description of the identified topics and helps contextualize technical relationships.

Relevance Assessment and Trend Analysis

For identified topics, the solution calculates quantitative relevance scores and visualizes their development over extended periods of time. This reveals which topics are gaining importance, which are losing relevance, and which trends are becoming established.

The analysis follows a consistent methodology. This improves the comparability of recurring assessments and reduces reliance on individual manual evaluations.

Interactive Access and Traceability

An interactive front end allows users to explore identified topics and trends. Users can navigate from an overview to individual news articles, view AI-generated summaries, and check the underlying sources.

In addition, the platform regularly generates overviews of the most important topics and subtopics within a given time period. This enables departments to prepare monthly reviews or management updates more efficiently.

Customer Benefits

Identify relevant pharmaceutical trends earlier. 

AI-powered news analysis helps departments continuously evaluate large volumes of pharmaceutical news, identify emerging topics, and provide a more reliable context for developments over time.

Faster

Analysis time reduced from days to minutes

The platform shortens the analysis process from several days to minutes. As a result, new trends and changes in the market become apparent much sooner.

More efficient

Saved several workdays per report

Automated data collection, analysis, and processing reduce the amount of manual research required. Several workdays are saved per report.

Can be used worldwide

Real-time competition information available

Industry sectors can directly review current topics, trends, and underlying sources. The platform was designed for global use and provides real-time competitive intelligence.

Our Strengths, Your Benefits

Analyzing Unstructured Pharmaceutical Data with AI

HMS combines data engineering, software engineering, and AI methodologies to create solutions that can be integrated into existing workflows. The project focused on developing a platform that continuously processes large volumes of unstructured news data and makes it analyzable and understandable for business units.

What You Can Expect from HMS

HMS combines data engineering, software engineering, and AI methodologies to create solutions that remain practically applicable and can be integrated into existing system landscapes.

  • Development of scalable data architectures for unstructured content
  • Combining machine learning, NLP, and generative AI for analytical use cases
  • Implementation of transparent analytical interfaces with access to data sources and individual results
  • Integration of AI solutions into existing system environments and business processes
Christoph Bergen
Christoph Bergen
CoE Lead GenAI

Sie möchten ein ähnliches Projekt umsetzen?

Sprechen Sie mit HMS darüber, wie sich Ihre Daten-, KI- oder Softwarelösung fachlich einordnen, technisch umsetzen und in Ihre bestehende Systemlandschaft integrieren lässt.

Projekt besprechen

Next Steps and Further Study

Data Science & Machine Learning

Learn how HMS uses machine learning, NLP, and generative AI to analyze unstructured data, identify relevant patterns, and make analytical applications productive.

More on Data Science & Machine Learning

Other Projects

Discover more case studies on generative AI, intelligent document analysis, AI assistants, and agentic AI.

Our Projects