nervaluate
March 9, 2026 · View on GitHub
nervaluate
nervaluate is a module for evaluating Named Entity Recognition (NER) models as defined in the SemEval 2013 - 9.1 task.
The evaluation metrics output by nervaluate go beyond a simple token/tag based schema, and consider different scenarios based on whether all the tokens that belong to a named entity were classified or not, and also whether the correct entity type was assigned.
This full problem is described in detail in the original blog post by David Batista, and this package extends the code in the original repository which accompanied the blog post.
The code draws heavily on the papers:
Usage example
pip install nervaluate
A possible input format are lists of NER labels, where each list corresponds to a sentence and each label is a token label.
Initialize the Evaluator class with the true labels and predicted labels, and specify the entity types we want to evaluate.
from nervaluate.evaluator import Evaluator
true = [
['O', 'B-PER', 'I-PER', 'O', 'O', 'O', 'B-ORG', 'I-ORG'], # "The John Smith who works at Google Inc"
['O', 'B-LOC', 'B-PER', 'I-PER', 'O', 'O', 'B-DATE'], # "In Paris Marie Curie lived in 1895"
]
pred = [
['O', 'O', 'B-PER', 'I-PER', 'O', 'O', 'B-ORG', 'I-ORG'],
['O', 'B-LOC', 'I-LOC', 'B-PER', 'O', 'O', 'B-DATE'],
]
evaluator = Evaluator(true, pred, tags=['PER', 'ORG', 'LOC', 'DATE'], loader="list")
Print the summary report for the evaluation, which will show the metrics for each entity type and evaluation scenario:
print(evaluator.summary_report())
Scenario: all
correct incorrect partial missed spurious precision recall f1-score
ent_type 5 0 0 0 0 1.00 1.00 1.00
exact 2 3 0 0 0 0.40 0.40 0.40
partial 2 0 3 0 0 0.70 0.70 0.70
strict 2 3 0 0 0 0.40 0.40 0.40
or aggregated by entity type under a specific evaluation scenario:
print(evaluator.summary_report(mode='entities'))
Scenario: strict
correct incorrect partial missed spurious precision recall f1-score
DATE 1 0 0 0 0 1.00 1.00 1.00
LOC 0 1 0 0 0 0.00 0.00 0.00
ORG 1 0 0 0 0 1.00 1.00 1.00
PER 0 2 0 0 0 0.00 0.00 0.00
Evaluation Scenarios
Token level evaluation for NER is too simplistic
When running machine learning models for NER, it is common to report metrics at the individual token level. This may not be the best approach, as a named entity can be made up of multiple tokens, so a full-entity accuracy would be desirable.
When comparing the golden standard annotations with the output of a NER system different scenarios might occur:
I. Surface string and entity type match
| Token | Gold | Prediction |
|---|---|---|
| in | O | O |
| New | B-LOC | B-LOC |
| York | I-LOC | I-LOC |
| . | O | O |
II. System hypothesized an incorrect entity
| Token | Gold | Prediction |
|---|---|---|
| an | O | O |
| Awful | O | B-ORG |
| Headache | O | I-ORG |
| in | O | O |
III. System misses an entity
| Token | Gold | Prediction |
|---|---|---|
| in | O | O |
| Palo | B-LOC | O |
| Alto | I-LOC | O |
| , | O | O |
Based on these three scenarios we have a simple classification evaluation that can be measured in terms of false positives, true positives, false negatives and false positives, and subsequently compute precision, recall and F1-score for each named-entity type.
However, this simple schema ignores the possibility of partial matches or other scenarios when the NER system gets the named-entity surface string correct but the type wrong. We might also want to evaluate these scenarios again at a full-entity level.
For example:
IV. System identifies the surface string but assigns the wrong entity type
| Token | Gold | Prediction |
|---|---|---|
| I | O | O |
| live | O | O |
| in | O | O |
| Palo | B-LOC | B-ORG |
| Alto | I-LOC | I-ORG |
| , | O | O |
V. System gets the boundaries of the surface string wrong
| Token | Gold | Prediction |
|---|---|---|
| Unless | O | B-PER |
| Karl | B-PER | I-PER |
| Smith | I-PER | I-PER |
| resigns | O | O |
VI. System gets the boundaries and entity type wrong
| Token | Gold | Prediction |
|---|---|---|
| Unless | O | B-ORG |
| Karl | B-PER | I-ORG |
| Smith | I-PER | I-ORG |
| resigns | O | O |
Defining evaluation metrics
How can we incorporate these described scenarios into evaluation metrics? See the original blog for a great explanation, a summary is included here.
We can define the following five metrics to consider different categories of errors:
| Error type | Explanation |
|---|---|
| Correct (COR) | both are the same |
| Incorrect (INC) | the output of a system and the golden annotation don’t match |
| Partial (PAR) | system and the golden annotation are somewhat “similar” but not the same |
| Missing (MIS) | a golden annotation is not captured by a system |
| Spurious (SPU) | system produces a response which doesn’t exist in the golden annotation |
These five metrics can be measured in four different ways:
| Evaluation schema | Explanation |
|---|---|
| Strict | exact boundary surface string match and entity type |
| Exact | exact boundary match over the surface string, regardless of the type |
| Partial | partial boundary match over the surface string, regardless of the type |
| Type | some overlap between the system tagged entity and the gold annotation is required |
These five errors and four evaluation schema interact in the following ways:
| Scenario | Gold entity | Gold string | Pred entity | Pred string | Type | Partial | Exact | Strict |
|---|---|---|---|---|---|---|---|---|
| III | BRAND | tikosyn | MIS | MIS | MIS | MIS | ||
| II | BRAND | healthy | SPU | SPU | SPU | SPU | ||
| V | DRUG | warfarin | DRUG | of warfarin | COR | PAR | INC | INC |
| IV | DRUG | propranolol | BRAND | propranolol | INC | COR | COR | INC |
| I | DRUG | phenytoin | DRUG | phenytoin | COR | COR | COR | COR |
| VI | GROUP | contraceptives | DRUG | oral contraceptives | INC | PAR | INC | INC |
Then precision, recall and f1-score are calculated for each different evaluation schema. In order to achieve data, two more quantities need to be calculated:
POSSIBLE (POS) = COR + INC + PAR + MIS = TP + FN
ACTUAL (ACT) = COR + INC + PAR + SPU = TP + FP
Then we can compute precision, recall, f1-score, where roughly describing precision is the percentage of correct named-entities found by the NER system. Recall as the percentage of the named-entities in the golden annotations that are retrieved by the NER system.
This is computed in two different ways depending on whether we want an exact match (i.e., strict and exact ) or a partial match (i.e., partial and type) scenario:
Exact Match (i.e., strict and exact )
Precision = (COR / ACT) = TP / (TP + FP)
Recall = (COR / POS) = TP / (TP+FN)
Partial Match (i.e., partial and type)
Precision = (COR + 0.5 × PAR) / ACT = TP / (TP + FP)
Recall = (COR + 0.5 × PAR)/POS = COR / ACT = TP / (TP + FN)
Putting all together:
| Measure | Type | Partial | Exact | Strict |
|---|---|---|---|---|
| Correct | 3 | 3 | 3 | 2 |
| Incorrect | 2 | 0 | 2 | 3 |
| Partial | 0 | 2 | 0 | 0 |
| Missed | 1 | 1 | 1 | 1 |
| Spurious | 1 | 1 | 1 | 1 |
| Precision | 0.5 | 0.66 | 0.5 | 0.33 |
| Recall | 0.5 | 0.66 | 0.5 | 0.33 |
| F1 | 0.5 | 0.66 | 0.5 | 0.33 |
Notes:
In scenarios IV and VI the entity type of the true and pred does not match, in both cases we only scored against
the true entity, not the predicted one. You can argue that the predicted entity could also be scored as spurious,
but according to the definition of spurious:
- Spurious (SPU) : system produces a response which does not exist in the golden annotation;
In this case there exists an annotation, but with a different entity type, so we assume it's only incorrect.
For the Type (ent_type) strategy, if multiple true entities of the same label overlap a
prediction, the match is resolved by closest boundaries. This can change which
(instance_index, entity_index) appears in missed_indices compared to list order,
while aggregate counts stay the same.
Contributing to the nervaluate package
Extending the package to accept more formats
The Evaluator accepts the following formats:
- Nested lists containing NER labels
- CoNLL style tab delimited strings
- prodi.gy style lists of spans
Additional formats can easily be added by creating a new loader class in nervaluate/loaders.py. The loader class
should inherit from the DataLoader base class and implement the load method.
The load method should return a list of entity lists, where each entity is represented as a dictionary
with label, start, and end keys.
The new loader can then be added to the _setup_loaders method in the Evaluator class, and can be selected with the
loader argument when instantiating the Evaluator class.
Here is list of formats we intend to include.
General Contributing
Improvements, adding new features and bug fixes are welcome. If you wish to participate in the development of nervaluate
please read the guidelines in the CONTRIBUTING.md file.
Give a ⭐️ if this project helped you!