Tools.

Software, frameworks and reusable research resources from DatAI. Find the project website and go straight to the source code.

Semantic data & knowledge graphs

Enrich tables, link entities and explore the structure of knowledge graphs.

  • SemT-X

    Framework

    A framework for semantic enrichment of tabular data, linking cells and columns to external knowledge sources through an interactive workspace.

  • MantisTable UI

    Interface

    A web interface for managing semantic table interpretation workflows and interacting with annotated tabular data.

  • LamAPI

    Service

    Knowledge-graph indexing and entity retrieval with string matching, relations and type-based filtering.

  • Alligator

    Library

    A Python library for entity linking over tabular data, with multiple annotation strategies and optional caching.

  • s-elBat

    Tool

    Entity retrieval and disambiguation for semantic table interpretation, supporting supervised and unsupervised workflows.

  • stEELlm

    Model

    An approach to cell entity annotation that uses a fine-tuned language model to generate semantic annotations of tabular data.

  • ABSTAT

    Framework

    Knowledge-graph profiling through compact patterns and statistics for understanding, exploration and data-quality assessment.

  • MammoTab25

    Dataset & code

    A semantic table interpretation dataset and supporting code for extracting, cleaning and annotating Wikipedia tables.

Document analysis & reasoning

Explore document collections, express analytical tasks and reason over factual knowledge.

  • DAVE

    Framework

    A human-in-the-loop workspace for entity-centred exploration, faceted search, conversational querying and knowledge refinement in document collections.

  • VCL

    Language & tool

    Verbal Command Language translates document-analysis requests into structured operations, execution plans and tasks over a document corpus.

  • ReFactX

    Method & code

    A method for reasoning with factual knowledge through constrained generation and an index of verbalised knowledge-graph facts.

Data pipelines & distributed systems

Build enrichment workflows and coordinate computation across distributed infrastructure.

  • Prompt2DAG

    Method & code

    A modular prompting approach for generating data-orchestration DAGs from natural-language pipeline descriptions.

  • SemT-py

    Library

    The Python library in the SemT enrichment ecosystem for transforming tables, reconciling values and extending datasets with external information.

  • DFaaS

    Platform

    A decentralised Function-as-a-Service platform that uses peer-to-peer coordination to distribute workloads across federated edge nodes.

Applied research prototypes

Explore software developed for specific research and application settings.

  • LLMind Chat

    Prototype

    A clinician-facing research prototype for exploring psychiatric case analysis with retrieval-augmented language models and expert evaluation.

  • Eventour

    Framework

    A research framework combining enriched urban knowledge graphs, language models and personalised walking itineraries for large-scale events.

Follow the research questions, projects and publications behind these tools.

Explore our research