Averbis provides text mining and natural language processing solutions including terminology management, text classification, and information extraction for domains such as life sciences, automotive, and publishing. They were awarded a contract by the European Patent Office to develop an automated patent classification system using text mining to classify over 80 million unpublished and published patents into the over 250,000 patent classification codes. The new system aimed to improve classification quality and speed compared to the previous manual process.

1. David Baehrens
Large-Scale Patent Classification
at the European Patent Office

2. ABOUT AVERBIS
Founded: 2007
Location: Freiburg im Breisgau
Team: Domain & IT-Experts
Focus: Leverage structured & unstructured information
Current Sectors: Pharma, Health, Automotive, Publishers & Libraries

3. PORTFOLIO
Solutions
Libraries PharmaPatentsHealthcare Social Media
Terminology
Management Text Mining
Search &
Analytics NoSQL
Categorization
& Clustering
Automotive

4. TERMINOLOGY MANAGEMENT
Terminology management
software
Provision of terminologies
Mappings between
terminologies
Building terminology-based
applications

5. Synonyms: dimethyl sulfoxide, dimethylsulfoxide, Domoso, Infiltrina
Hierarchies: cancer, carcinoma, melanoma, lymphoma, glioblastoma…
Patterns: dates, citations, mail addresses…
Rule-based extraction of all different kinds of complex information
Persons, Locations, Genes, ….
Coocurrences, Typed Relations, e.g. Genes / Diseases / Modification Type
TEXT MINING
Term Detection
Regular
Expressions
Rule Engine
Named Entities
Relations
Sentences, Tokens, POS-Tags, Chunks, Paragraphs, Sections, Stemming, Decompounding…Syntax Detection

6. RULE ENGINE
1. NAME OF THE MEDICINAL PRODUCT
Desloratadine ratiopharm 5 mg film-coated tablets
Primary Field Name Secondary Field Name Field Value
MedicalProductName coveredText Desloratadine ratiopharm 5 mg film-coated tablets
inventedPartName DESLORATADINE
strengthPart 5 mg
pharmaceuticalDoseFormPart FILM-COATED TABLET
TextRegelErgebnis

7. SEARCH & NOSQL
Free text + concept based
search
Text mining integration
Guided navigation / facets
NoSQL functionalities
Multi- & cross lingual search
Related documents
Based on Apache Solr
• Extended Query Syntax
• JSON-API
• Scalability
…

8. DOCUMENT CLASSIFICATION
Hotel Reviews
Patents

9. SEARCH & NOSQL

10. INFORMATION DISCOVERY
Terminology
Management Text Mining
Search &
Analytics NoSQL
Categorization
& Clustering
Delivery / Deployment / Runtime Environment
Integration Tests / Continuous Integration
Extensive Documentation
Common Architecture / Application Design
User & Role Management, Security
Communication Bus
Project Management

11. PATENT CLASSIFICATION AT EPO
Tender No. 1585
1) Pre-Classification of
unpublished patents into departments
2) Re-Classification on
published patents, if category system changes

12. ABOUT EPO
• The European Patent Office (EPO)
grants European patents for the
Contracting States to the European
Patent Convention
• Second largest intergovernmental
institution in Europe
• Not an EU institution
• Self-financing, i.e. revenue
from fees covers operating
and capital expenditure

13. NUMBER OF STAFF
Status: December 2008

14. PATENT APPLICATIONS

15. http://www.epo.org/about-us/annual-reports-statistics/annual-report/2014.html

16. COOPERATIVE PATENT CLASSIFICATION
• Patent Classification System based on ECLA / IPC
• jointly developed by the European Patent Office (EPO)
and the United States Patent and Trademark Office
(USPTO)
• used by both the EPO and USPTO since 1 January 2013
• currently contains about 250.000 classes

17. EXAMPLE CPC CLASS

18. GRANTED PATENT

19. EARLY PATENT

20. EARLY PATENT

21. EARLY PATENT

22. PATENT CLASSIFICATION AT EPO
Tender No. 1585
1) Pre-Classification of
unpublished patents into departments
Our Motivation:
• Great Classification Use-Case
– Big Data (80 Mio. patents available)
– Large Scale Category System >250.000 CPC codes
– Tough classification quality and response time
constraints
• Text Mining Success Story

23. OLD CLASSIFICATION PROCESS
PATENTS CLA SSIFICATION DEPARTMENTS

24. CLASSIFICATION COMPLEXITY
~250.000
CPC Codes
~1.500
Ranges
250
Departments

25. CLASSIFICATION PROCESS
PATENTS CLA SSIFICATION DEPARTMENTS

26. NEW CLASSIFICATION PROCESS
PATENTS CLA SSIFICATION DEPARTMENTS

27. SOME FACTS
• about 650k training documents from 2005-2013
• supervised learning: light-weight and fast linear support
vector machine
• Training time (16 Cores, 128 GB RAM)
– Feature Extraction: ~1 hour
– Training of Classifiers: ~1 hour
– 90/10 tests with a look-a-head of 3 levels
and reporting 3 best candidates: ~1 hour
• Prediction: 5 docs in 5 sec

28. HIERARCHICAL CLASSIFICATION

29. STATUS & OUTLOOK
 Range-specific quality
evaluation
 Going live with best
ranges
• Continuous optimization

30. PATENT CLASSIFICATION AT EPO
Tender No. 1585
1) Re-Classification on
published patents, if category system changes
Challenges and Facts:
– 250.000 CPC codes, regular changes/refinements
– Several re-classification projects at any one time, great
variation in size, a class is split into 5-20(?) subclasses
– No training material available

31. NEW RE-CLASSIFICATION PROCESS
Training Data
• Human Annotator starts labeling about 20% of
the documents with new subclasses
Statistical Models
• are generated on-the-fly, and
• Cross-validation test are carried out
Threshold
• If cross-validation achieves certain threshold
(e.g. 90%), the remaining documents are
classified fully automatically without further
review
• Otherwise, more training data is being generated

32. STATUS & OUTLOOK
 Currently in evaluation
phase
• Going live in the next
weeks

33. …NOT ONLY PATENTS
Solutions
Libraries PharmaPatentsHealthcare Social Media
Terminology
Management Text Mining
Search &
Analytics NoSQL
Categorization
& Clustering
Automotive

34. For further questions, please contact:
David Baehrens
 + 49 (0)761 203 97690
 info@averbis.com