extractacy 0.1.0

A SpaCy pipeline object for extracting values that correspond to a named entity (e.g., birth dates, account numbers, or laboratory results)

extractacy - value extraction and linking for spaCy

spaCy pipeline object for extracting values that correspond to a named entity (e.g., birth dates, account numbers, or laboratory results)

Installation and usage

Install the library.

pip install extractacy

Import library and spaCy.

import spacy
from extractacy.extract import ValueExtractor

Load spacy language model. Set up an EntityRuler for the example. Define the entites and value extraction patterns and add to nlp pipeline.

nlp = spacy.load("en_core_web_sm")
# Set up entity ruler
ruler = EntityRuler(nlp)
patterns = [
    {"label": "TEMP_READING", "pattern": [{"LOWER": "temperature"}]},
    {"label": "TEMP_READING", "pattern": [{"LOWER": "temp"}]},
    {
        "label": "DISCHARGE_DATE",
        "pattern": [{"LOWER": "discharge"}, {"LOWER": "date"}],
    },
    
]
ruler.add_patterns(patterns)
nlp.add_pipe(ruler, last=True)

# Define ent_patterns for value extraction
ent_patterns = {
    "DISCHARGE_DATE": {"patterns": [[{"SHAPE": "dd/dd/dddd"}, {"SHAPE": "dd/d/dddd"}]],"n": 2, "direction": "right"},
    "TEMP_READING": {"patterns": [[
                        {"LIKE_NUM": True},
                        {"LOWER": {"IN": ["f", "c", "farenheit", "celcius", "centigrade", "degrees"]}
                        },
                    ]
                ],
                "n": "sent",
                "direction": "both"
        },
}

valext = ValueExtractor(nlp, ent_patterns)
nlp.add_pipe(valext, last=True)

doc = nlp("Discharge Date: 11/15/2008. Patient had temp reading of 102.6 degrees.")
for e in doc.ents:
    if e._.value_extract:
        print(e.text, e.label_, e._.value_extract)
## Discharge Date DISCHARGE_DATE 11/15/2008
## temp reading TEMP_READING 102.6 degrees

Value Extraction patterns

Returns all patterns within n tokens of entity of interest or within the same sentence. It relies on spaCy token matching syntax.

{"ENTITY_NAME":{"patterns":[{"LOWER":"awesome"}, {"LOWER":"pattern"}], "n": 5, "direction": "right"}}

Use "n":"sent" for within sentence method rather than n tokens.

Contributing

contributing

Authors

• Jeno Pizarro

License

license