Enum: AgentTypeEnum

Permissible Values

Value	Meaning	Description
manual_agent	None	A human agent who is responsible for generating a statement of knowledge
automated_agent	None	An automated agent, typically a software program or tool, that is responsibl...
data_analysis_pipeline	None	An automated agent that executes an analysis workflow over data and reports ...
computational_model	None	An automated agent that generates knowledge statements (typically predictions...
text_mining_agent	None	An automated agent that uses Natural Language Processing to recognize concept...
image_processing_agent	None	An automated agent that processes images to generate textual statements of k...
manual_validation_of_automated_agent	None	A human agent reviews and validates/approves the veracity of knowledge that ...
not_provided	None	The agent type is not provided, typically because it cannot be determined fro...

Slots

Name	Description
agent_type	Type of agent involved in the relationship

Identifier and Mapping Information

Schema Source

from schema: https://w3id.org/everycure-org/matrix-schema

LinkML Source

name: AgentTypeEnum
from_schema: https://w3id.org/everycure-org/matrix-schema
rank: 1000
permissible_values:
  manual_agent:
    text: manual_agent
    description: A human agent who is responsible for generating a statement of knowledge.
      The human may utilize computationally generated information as evidence for
      the resulting knowledge,  but the human is the one who ultimately interprets/reasons
      with  this evidence to produce a statement of knowledge.
  automated_agent:
    text: automated_agent
    description: An automated agent, typically a software program or tool, that is  responsible
      for generating a statement of knowledge. Human contribution  to the knowledge
      creation process ends with the definition and coding of algorithms or analysis
      pipelines that get executed by the automated agent.
  data_analysis_pipeline:
    text: data_analysis_pipeline
    description: An automated agent that executes an analysis workflow over data and  reports
      the direct results of the analysis. These typically report  statistical associations/correlations
      between variables in the input dataset, and do not interpret/infer broader conclusions
      from associations the analysis reveals in the data.
    is_a: automated_agent
    notes:
    - If an analysis pipeline includes any rules for generating broader  conclusions
      based on the dataset-specific statistical correlations it calculates (e.g. create
      a 'treats' edge when the analysis reveals a  drug-disease correlation in the
      data with statistical scores that meet a  certain threshold) - we would consider
      this agent to be a Computational Model rather than just a Data Analysis Pipeline.
  computational_model:
    text: computational_model
    description: An automated agent that generates knowledge statements (typically
      predictions) based on rules/logic explicitly encoded in an algorithm (e.g. heuristic
      models, supervised classifiers), or learned from patterns  observed in data
      (e.g. ML models, unsupervised classifiers).
    is_a: automated_agent
    notes:
    - The bar is quite low relatively for what is considered to be a  ‘computational
      model’ by our definition. Even agents/tools that apply  simple rules or logic
      to the output of an ingest or analysis pipeline to allow for a stronger or more
      general conclusion to be stated can  qualify an agent as a model. For example,
      an ingest pipeline that applies rules to its ingest of  clinical trials data
      to create a 'treats' prediction edge when the  source reports a drug to be in
      phase 2 or 3 trials represents a computational model because it is automatically
      drawing a stronger conclusion than the source reports, based on logic encoded
      in the ingest pipeline. Similarly, a data analysis pipeline that is extended
      with rules to  automatically generate broader conclusions based on dataset-specific
      statistical correlations (e.g. create a 'treats' edge when the analysis reveals
      a drug-disease correlation in the data with statistical scores  that meet a
      certain threshold), would also qualify as a computational  model by our definition.
  text_mining_agent:
    text: text_mining_agent
    description: An automated agent that uses Natural Language Processing to recognize
      concepts and/or relationships in text, and report them using formally encoded
      semantics (e.g. as an edge in a knowledge graph).
    is_a: automated_agent
    notes:
    - The original statement in the source text is typically made by a human /  manual
      agent, but if a specific encoding of this knowledge is produced by a text-mining
      tool, it has an agent_type of 'text_mining_agent'. Examples of text mining agents
      include SemmedDB, and the Translator Text-Mining Knowledge Provider. Note that
      text-mining tools are prone to erroneous interpretation of  concepts and relationships,
      and can fail to provide important details  about the context in which the original
      knowledge was reported - so users should always consult the source text for
      a text-mined statement to assess its veracity and relevance.
  image_processing_agent:
    text: image_processing_agent
    description: An automated agent that processes images to generate textual statements
      of  knowledge derived from the image and/or expressed in text the image  depicts
      (e.g. via OCR).
    is_a: automated_agent
  manual_validation_of_automated_agent:
    text: manual_validation_of_automated_agent
    description: A human agent reviews and validates/approves the veracity of knowledge  that
      is initially generated by an automated agent.
    notes:
    - This term applies when a human was only involved in evaluating the veracity
      of a knowledge statement that was generated by an automated agent. It is  important
      to indicate when such manual review has occurred, because it can give a user
      more confidence in an automated statement.
  not_provided:
    text: not_provided
    description: The agent type is not provided, typically because it cannot be determined
      from available information if the agent that generated the knowledge is  manual
      or automated.