--- # For reference on model card metadata, see the spec: https://github.com/huggingface/hub-docs/blob/main/modelcard.md?plain=1 # Doc / guide: https://huggingface.co/docs/hub/model-cards {} --- # Patent Entity Extraction Model ### Model Description **patent_entities_ner** is a fine-tuned [XLM-RoBERTa-large](https://huggingface.co/FacebookAI/xlm-roberta-large) model that has been trained on a custom dataset of OCR'd front pages of patent specifications published by the British Patent Office, and filed between 1617-1899. It has been trained to recognize six classes of named entities: - PER: full name of inventor - OCC: occupation of inventor - ADD: full (permanent) address of inventor - DATE: patent filing, submission, or approval dates - FIRM: name of firm affiliated with inventor - COMM: name and information mentioned about communicant We take the original xlm-roberta-large [weights](https://huggingface.co/FacebookAI/xlm-roberta-large/blob/main/pytorch_model.bin) and fine tune on our custom dataset for 29 epochs with a learning rate of 5e-05 and a batch size of 21. We chose the learning rate by tuning on the validation set. ### Usage This model can be used with HuggingFace Transformer's Pipelines API for NER: ```python from transformers import pipeline, AutoModelForTokenClassification, AutoTokenizer tokenizer = AutoTokenizer.from_pretrained("gbpatentdata/patent_entities_ner") model = AutoModelForTokenClassification.from_pretrained("gbpatentdata/patent_entities_ner") def custom_recognizer(text, model=model, tokenizer=tokenizer, device=0): # HF ner pipeline token_level_results = pipeline("ner", model=model, device=0, tokenizer=tokenizer)(text) # keep entities tracked entities = [] current_entity = None for item in token_level_results: tag = item['entity'] # replace '▁' with space for easier reading (_ is created by the XLM-RoBERTa tokenizer) word = item['word'].replace('▁', ' ') # aggregate I-O-B tagged entities if tag.startswith('B-'): if current_entity: entities.append(current_entity) current_entity = {'type': tag[2:], 'text': word.strip(), 'start': item['start'], 'end': item['end']} elif tag.startswith('I-'): if current_entity and tag[2:] == current_entity['type']: current_entity['text'] += word current_entity['end'] = item['end'] else: if current_entity: entities.append(current_entity) current_entity = {'type': tag[2:], 'text': word.strip(), 'start': item['start'], 'end': item['end']} else: # deal with O tag if current_entity: entities.append(current_entity) current_entity = None if current_entity: # add to entities entities.append(current_entity) # track entity merges merged_entities = [] # merge entities of the same type for entity in entities: if merged_entities and merged_entities[-1]['type'] == entity['type'] and merged_entities[-1]['end'] == entity['start']: merged_entities[-1]['text'] += entity['text'] merged_entities[-1]['end'] = entity['end'] else: merged_entities.append(entity) # clean up extra spaces for entity in merged_entities: entity['text'] = ' '.join(entity['text'].split()) # convert to list of dicts return [{'class': entity['type'], 'entity_text': entity['text'], 'start': entity['start'], 'end': entity['end']} for entity in merged_entities] example = """ Date of Application, 1st Aug., 1890-Accepted, 6th Sept., 1890 COMPLETE SPECIFICATION. Improvements in Coin-freed Apparatus for the Sale of Goods. I, CHARLES LOTINGA, of 33 Cambridge Street, Lower Grange, Cardiff, in the County of Glamorgan, Gentleman, do hereby declare the nature of this invention and in what manner the same is to be performed, to be particularly described and ascertained in and by the following statement """ ner_results = custom_recognizer(example) print(ner_results) ``` ### Training Data The custom dataset of front page texts of patent specifications was assembled in the following steps: 1. We fine tuned a YOLO vision [model](https://huggingface.co/gbpatentdata/yolov8_patent_layouts) to detect bounding boxes around text. We use this to identify text regions on the front pages of patent specifications. 2. We use [Google Cloud Vision](https://cloud.google.com/vision?hl=en) to OCR the detected text regions, and then concatenate the OCR text. 3. We randomly sample 200 front page texts (and another 201 oversampled from those that contain either firm or communicant information). Our custom dataset has accurate manual labels created jointly by an undergraduate student and an economics professor. The final dataset is split 60-20-20 (train-val-test). In the event that the front page text is too long, we restrict the text to the first 512 tokens. ### Evaluation Our evaluation metric is F1 at the full entity-level. That is, we aggregated adjacent-indexed entities into full entities and computed F1 scores requiring an exact match. These scores for the test set are below.
Full Entity | Precision | Recall | F1-Score |
---|---|---|---|
PER | 92.2% | 97.7% | 94.9% |
OCC | 93.8% | 93.8% | 93.8% |
ADD | 88.6% | 91.2% | 89.9% |
DATE | 93.7% | 98.7% | 96.1% |
FIRM | 64.0% | 94.1% | 76.2% |
COMM | 77.1% | 87.1% | 81.8% |
Overall (micro avg) | 89.9% | 95.3% | 92.5% |
Overall (macro avg) | 84.9% | 93.8% | 88.9% |
Overall (weighted avg) | 90.3% | 95.3% | 92.7% |