Skip to main content
Loading navigation
Back to challenges

Genomics Executive Leadership & Personnel Intelligence Engine with spaCy and duckdb

Tracking executive movements and corporate strategy across biotech firms requires structured entity linking from industry news feeds. You will build an NLP pipeline using spaCy and DuckDB to extract personnel appointments and corporate leadership shifts, reaching an extraction F1-score of 0.90.

Data ScienceHosted by Vera
Sign in free to start

What you are building

Develop an NLP and analytical database pipeline using spaCy and DuckDB to aggregate executive transitions in genomics.

How work is evaluated

Evaluation

Evaluates named entity recognition F1-score and DuckDB query accuracy on executive transition news text.

Datasets

Public datasets, plus the private datasets you have access to.

How submissions are scored

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

entity_match_check

Ensure executive and company names match ground truth

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

f1_score

F1-score of named entity recognition on corporate movement text • target: 0.9 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

  • Build custom Named Entity Recognition (NER) models with spaCy for executive titles and company names

  • Index structured entity records in high-performance DuckDB analytical tables

  • Analyze executive hiring trends across sequencing and spatial biology sectors

Resources

Dataset notes

Dataset of 200 biopharma executive press releases and news snippets with annotated entities.

Challenge setup

Evaluates named entity recognition F1-score and DuckDB query accuracy on executive transition news text.

Challenge input

JSON object with article_text string

spaCy

Industrial natural language processing toolkit.

DuckDB

Fast in-process relational database.

Evaluated output

JSON object with executive_name string, company string, and role string

Evaluation checks
  • Ensure executive and company names match ground truth
  • F1-score of named entity recognition on corporate movement text • target: 0.9 • range: 0-1
Configured evaluation evidence
  • Benchmark: BioNER-Bench
  • F1 Score target: 0.9
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Shows the first task, up to three tools, and up to three evidence items.

View technical recipe

Configured tools

Action Space
  • Graphite · Required
  • QA Wolf · Optional
  • Mistral · Optional

Evaluation contract

  • entity_match_check · Weight 1
  • f1_score · Weight 1

Metrics

  • f1_score · Target 0.9

Benchmarks

  • BioNER-Bench
  • general

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON object with article_text string

    Output: JSON object with executive_name string, company string, and role string

    Execution harness: Python

Evaluation tools

  • spaCy · NLP Framework
  • DuckDB · Analytical Database Engine

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Genomics Executive Leadership & Personnel Intelligence Engine with spaCy and duckdb