Luxid® Skill Cartridge® Library

Luxid® Skill Cartridge® Library is a range of off-the shelf extractors developed by TEMIS to address recurring use cases and domains



Here is an overview of some of the most representative of these Skill Cartridges®.


Relevant Term Finder Skill Cartridge® (RTF)

The RTF Skill Cartridge® identifies the most characteristic terms of each document (its "fingerprint") by comparing its vocabulary to a statistical model based on a reference corpus. This model can be adjusted through training on any corpus that appropriately reflects your specific domain. RTF requires no pre-defined conceptual structure and can be successfully applied to use cases such as Similar Documents Recommendation, Corpus Clustering, as well as Domain-Specific Terminology Extraction.

Smart Taxonomy Facilitator Skill Cartridge® (STF)

STF acts as a vehicle for applying taxonomies and controlled vocabularies to documents. It embeds technologies that help overcome two key weaknesses associated to taxonomy-based indexing. The first, Fuzzy Term Matching, automatically produces variants of the forms present in the taxonomy, thereby helping to improve recall. The second, Relevance Scoring, applies a range of heuristics to assign a relevance score to each extracted concept and discards the less relevant ones, therefore improving extraction precision. STF can also exploit part-of-speech tagging information to avoid false positives caused by ambiguous taxonomical terms. Luxid® Webstudio enables you to seamlessly package your own taxonomy into STF and adjust its performance to your specific use case. 

Machine Learning Extraction Skill Cartridge® (MLX)

The MLX Skill Cartridge® is a versatile extractor that can be trained to extract, subcategorize and/or score virtually any type of entity in text based on a previously annotated corpus where such entities of interest have been highlighted by domain experts. The corpus annotation and Skill Cartridge® training processes requires no natural language processing expertise and can be performed easily with Annotation Workbench. 

IPTC Categorization Skill Cartridge®

This Skill Cartridge® analyzes the vocabulary used in documents and classifies them into approximately 130 categories based on key headings of the top two levels of the IPTC (News) taxonomy. This Skill Cartridge® is available for the English and French languages. Extensions to other languages and more complete coverage of the IPTC headings can be achieved on a project basis.

Text Mining 360° Skill Cartridge® (TM360)

TM360 extracts more than 20 types of common entities among which the names of People, Companies, Organizations, Locations, as well as Measurements, Money and Time expressions, and Contact information. To achieve this task, TM360 embeds a variety of predictive natural language processing methods including morphological and syntactical heuristics. This Skill Cartridge® is available in eight languages (English, German, French, Spanish, Italian, Dutch, Portuguese and Arabic).

Competitive Intelligence Skill Cartridge® (CI)

CI extracts the same entities as the TM360° Skill Cartridge®, as well as 10 types of semantic relationships involving companies, their employees, assets or strategy, including Corporate Mergers and Acquisitions, Court Cases, Board and Management Changes, Financial Reporting, Business Development and Strategy. CI leverages syntactical reasoning to establish the role played by each entity in the relationship, enabling it to recognize for instance the acquiring company from the target or the licensee from the licensor. CI is available in English, French, Spanish and Dutch and can be customized through the use of specific lexicons as well as numerous parameters adjusting its extraction behavior.

Opinion Mining Skill Cartridge® (OM)

The Opinion Mining Skill Cartridge® recognizes the subjective expressions that reveal evaluative judgment or emotional states that are typically found in social media or customer feedback corpora. Leveraging a combination of specialized thesauri and syntactical reasoning, OM qualifies these opinions both in terms of polarity (negative/positive) and intensity, and links them to their target (for example products, service features, brands, organizations or people), helping analysts quantify, track and analyze information that is otherwise entirely unstructured. This Skill Cartridge® is available in English and French.

Biological Entities and Relationship Skill Cartridge® (BER)

Available in the English language, BER extracts both 15 key types of biological entities and 17 relationships binding them, as well as rich contextual information. Extracted entities include Anatomical terms, Cells and Tissues, Proteins and Genes, Disorders, Treatments, Species, Genomic and Mutational information. Relationships include Activation, Inhibition, Regulation, Gene expression, Mutation, Diagnosis, and Therapy. BER is powered by a deep syntactical analysis layer that provides an intimate and highly structured understanding of complex biological mechanisms and pathways.

BER has a modular structure that offers unique customization capabilities, enabling the assembly of its Skill Units into bespoke extractors focused on novel areas of interest.

Medical Entities and Relationships Skill Cartridge® (MER)

MER extracts 15 key types of medical entities and 10 types of semantic relationships binding them. Extracted entities include Clinical Trial terms, Diagnostics, Disorders, Transmission information, Genomic information, Species information, Ethnic information, Symptoms, Targets, Treatments, and Administration Routes.  Examples of relationships include Adverse Effects (binding a treatment with a negative effect), Therapies (binding a treatment with a disease), and Molecular Targets (binding a process with certain specific entity types).

MER refers to the MeSH nomenclature and can be customized with specialized or in-house terminology. MER is available in the English language.

Chemical Entities Recognition Skill Cartridge® (CER)

CER extracts chemical entities such as chemical compounds, chemical classes and molecular formulae and performs name-to-structure conversion. It includes support for International Union of Pure and Applied Chemistry (IUPAC) as well as Chemical Abstracts Service (CAS) registry numbers.

CER can be successfully used in conjunction with BER or MER to reveal the chemical compounds at play in biological and medical contexts. It can also be customized with custom lexicons. 


Beyond the above, Luxid® Skill Cartridge® Library also includes a range of Skill Units that are more specific in scope or application areas and can provide value in specific domains such as Military, Political and Strategic events, Legal proceedings, Scientific and Technical domains beyond the Life Sciences.

Related downloads
171 Mb
628 Kb
Structure, manage and exploit your unstructured content.