|
|
--- |
|
|
base_model: sentence-transformers/all-mpnet-base-v2 |
|
|
library_name: sentence-transformers |
|
|
pipeline_tag: sentence-similarity |
|
|
tags: |
|
|
- sentence-transformers |
|
|
- sentence-similarity |
|
|
- feature-extraction |
|
|
- generated_from_trainer |
|
|
- dataset_size:5579240 |
|
|
- loss:CachedMultipleNegativesRankingLoss |
|
|
widget: |
|
|
- source_sentence: Program Coordinator RN |
|
|
sentences: |
|
|
- >- |
|
|
discuss the medical history of the healthcare user, evidence-based approach |
|
|
in general practice, apply various lifting techniques, establish daily |
|
|
priorities, manage time, demonstrate disciplinary expertise, tolerate |
|
|
sitting for long periods, think critically, provide professional care in |
|
|
nursing, attend meetings, represent union members, nursing science, manage a |
|
|
multidisciplinary team involved in patient care, implement nursing care, |
|
|
customer service, work under supervision in care, keep up-to-date with |
|
|
training subjects, evidence-based nursing care, operate lifting equipment, |
|
|
follow code of ethics for biomedical practices, coordinate care, provide |
|
|
learning support in healthcare |
|
|
- >- |
|
|
provide written content, prepare visual data, design computer network, |
|
|
deliver visual presentation of data, communication, operate relational |
|
|
database management system, ICT communications protocols, document |
|
|
management, use threading techniques, search engines, computer science, |
|
|
analyse network bandwidth requirements, analyse network configuration and |
|
|
performance, develop architectural plans, conduct ICT code review, hardware |
|
|
architectures, computer engineering, video-games functionalities, conduct |
|
|
web searches, use databases, use online tools to collaborate |
|
|
- >- |
|
|
nursing science, administer appointments, administrative tasks in a medical |
|
|
environment, intravenous infusion, plan nursing care, prepare intravenous |
|
|
packs, work with nursing staff, supervise nursing staff, clinical perfusion |
|
|
- source_sentence: Director of Federal Business Development and Capture Mgmt |
|
|
sentences: |
|
|
- >- |
|
|
develop business plans, strive for company growth, develop personal skills, |
|
|
channel marketing, prepare financial projections, perform market research, |
|
|
identify new business opportunities, market research, maintain relationship |
|
|
with customers, manage government funding, achieve sales targets, build |
|
|
business relationships, expand the network of providers, make decisions, |
|
|
guarantee customer satisfaction, collaborate in the development of marketing |
|
|
strategies, analyse business plans, think analytically, develop revenue |
|
|
generation strategies, health care legislation, align efforts towards |
|
|
business development, assume responsibility, solve problems, deliver |
|
|
business research proposals, identify potential markets for companies |
|
|
- >- |
|
|
operate warehouse materials, goods transported from warehouse facilities, |
|
|
organise social work packages, coordinate orders from various suppliers, |
|
|
warehouse operations, work in assembly line teams, work in a logistics team, |
|
|
footwear materials |
|
|
- >- |
|
|
manufacturing plant equipment, use hand tools, assemble hardware components, |
|
|
use traditional toolbox tools, perform product testing, control panel |
|
|
components, perform pre-assembly quality checks, oversee equipment |
|
|
operation, assemble mechatronic units, arrange equipment repairs, assemble |
|
|
machines, build machines, resolve equipment malfunctions, electromechanics, |
|
|
develop assembly instructions, install hydraulic systems, revise quality |
|
|
control systems documentation, detect product defects, operate hydraulic |
|
|
machinery controls, show an exemplary leading role in an organisation, |
|
|
assemble manufactured pipeline parts, types of pallets, perform office |
|
|
routine activities, conform with production requirements, comply with |
|
|
quality standards related to healthcare practice |
|
|
- source_sentence: director of production |
|
|
sentences: |
|
|
- >- |
|
|
use customer relationship management software, sales strategies, create |
|
|
project specifications, document project progress, attend trade fairs, |
|
|
building automation, sales department processes, work independently, develop |
|
|
account strategy, build business relationships, facilitate the bidding |
|
|
process, close sales at auction, satisfy technical requirements, |
|
|
results-based management, achieve sales targets, manage sales teams, liaise |
|
|
with specialist contractors for well operations, sales activities, use sales |
|
|
forecasting softwares, guarantee customer satisfaction, integrate building |
|
|
requirements in the architectural design, participate actively in civic |
|
|
life, customer relationship management, implement sales strategies |
|
|
- >- |
|
|
translate strategy into operation, lead the brand strategic planning |
|
|
process, assist in developing marketing campaigns, implement sales |
|
|
strategies, sales promotion techniques, negotiate with employment agencies, |
|
|
perform market research, communicate with customers, develop media strategy, |
|
|
change power distribution systems, beverage products, project management, |
|
|
provide advertisement samples, devise military tactics, use microsoft |
|
|
office, market analysis, manage sales teams, create brand guidelines, brand |
|
|
marketing techniques, use sales forecasting softwares, supervise brand |
|
|
management, analyse packaging requirements, provide written content, hand |
|
|
out product samples, channel marketing |
|
|
- >- |
|
|
use microsoft office, use scripting programming, build team spirit, operate |
|
|
games, production processes, create project specifications, analyse |
|
|
production processes for improvement, manage production enterprise, Agile |
|
|
development, apply basic programming skills, document project progress, |
|
|
supervise game operations, work to develop physical ability to perform at |
|
|
the highest level in sport, fix meetings, office software, enhance |
|
|
production workflow, manage a team, set production KPI, manage commercial |
|
|
risks, work in teams, teamwork principles, address identified risks, meet |
|
|
deadlines, consult with production director |
|
|
- source_sentence: Nursing Assistant |
|
|
sentences: |
|
|
- >- |
|
|
supervise medical residents, observe healthcare users, provide domestic |
|
|
care, prepare health documentation, position patients undergoing |
|
|
interventions, work with broad variety of personalities, supervise food in |
|
|
healthcare, tend to elderly people, monitor patient's vital signs, transfer |
|
|
patients, show empathy, provide in-home support for disabled individuals, |
|
|
hygiene in a health care setting, supervise housekeeping operations, perform |
|
|
cleaning duties, monitor patient's health condition, provide basic support |
|
|
to patients, work with nursing staff, involve service users and carers in |
|
|
care planning, use electronic health records management system, arrange |
|
|
in-home services for patients, provide nursing care in community settings , |
|
|
work in shifts, supervise nursing staff |
|
|
- >- |
|
|
manage relationships with stakeholders, use microsoft office, maintain |
|
|
records of financial transactions, software components suppliers, tools for |
|
|
software configuration management, attend to detail, keep track of expenses, |
|
|
build business relationships, issue sales invoices, financial department |
|
|
processes, supplier management, process payments, perform records |
|
|
management, manage standard enterprise resource planning system |
|
|
- >- |
|
|
inspect quality of products, apply HACCP, test package, follow verbal |
|
|
instructions, laboratory equipment, assist in the production of laboratory |
|
|
documentation, ensure quality control in packaging, develop food safety |
|
|
programmes, packaging engineering, appropriate packaging of dangerous goods, |
|
|
maintain laboratory equipment, SAP Data Services, calibrate laboratory |
|
|
equipment, analyse packaging requirements, write English |
|
|
- source_sentence: Branch Manager |
|
|
sentences: |
|
|
- >- |
|
|
support employability of people with disabilities, schedule shifts, issue |
|
|
licences, funding methods, maintain correspondence records, computer |
|
|
equipment, decide on providing funds, tend filing machine, use microsoft |
|
|
office, lift stacks of paper, transport office equipment, tend to guests |
|
|
with special needs, provide written content, foreign affairs policy |
|
|
development, provide charity services, philanthropy, maintain financial |
|
|
records, meet deadlines, manage fundraising activities, assist individuals |
|
|
with disabilities in community activities, report on grants, prepare |
|
|
compliance documents, manage grant applications, tolerate sitting for long |
|
|
periods, follow work schedule |
|
|
- >- |
|
|
cook pastry products, create new recipes, food service operations, assess |
|
|
shelf life of food products, apply requirements concerning manufacturing of |
|
|
food and beverages, food waste monitoring systems, maintain work area |
|
|
cleanliness, comply with food safety and hygiene, coordinate catering, |
|
|
maintain store cleanliness, work according to recipe, health, safety and |
|
|
hygiene legislation, install refrigeration equipment, prepare desserts, |
|
|
measure precise food processing operations, conform with production |
|
|
requirements, work in an organised manner, demand excellence from |
|
|
performers, refrigerants, attend to detail, ensure food quality, manufacture |
|
|
prepared meals |
|
|
- >- |
|
|
teamwork principles, office administration, delegate responsibilities, |
|
|
create banking accounts, manage alarm system, make independent operating |
|
|
decisions, use microsoft office, offer financial services, ensure proper |
|
|
document management, own management skills, use spreadsheets software, |
|
|
manage cash flow, integrate community outreach, manage time, perform |
|
|
multiple tasks at the same time, carry out calculations, assess customer |
|
|
credibility, maintain customer service, team building, digitise documents, |
|
|
promote financial products, communication, assist customers, follow |
|
|
procedures in the event of an alarm, office equipment |
|
|
license: mit |
|
|
language: |
|
|
- en |
|
|
--- |
|
|
|
|
|
# SentenceTransformer based on sentence-transformers/all-mpnet-base-v2 |
|
|
|
|
|
This is a [sentence-transformers](https://www.SBERT.net) model specifically trained for job title matching and similarity. It's finetuned from [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) on a large dataset of job titles and their associated skills/requirements. The model maps job titles and descriptions to a 1024-dimensional dense vector space and can be used for semantic job title matching, job similarity search, and related HR/recruitment tasks. |
|
|
|
|
|
## Model Details |
|
|
|
|
|
### Model Description |
|
|
- **Model Type:** Sentence Transformer |
|
|
- **Base model:** [sentence-transformers/all-mpnet-base-v2](https://huggingface.co/sentence-transformers/all-mpnet-base-v2) |
|
|
- **Maximum Sequence Length:** 64 tokens |
|
|
- **Output Dimensionality:** 1024 tokens |
|
|
- **Similarity Function:** Cosine Similarity |
|
|
- **Training Dataset:** 5.5M+ job title - skills pairs |
|
|
- **Primary Use Case:** Job title matching and similarity |
|
|
- **Performance:** Achieves 0.6457 MAP on TalentCLEF benchmark |
|
|
|
|
|
### Model Sources |
|
|
|
|
|
- **Documentation:** [Sentence Transformers Documentation](https://sbert.net) |
|
|
- **Repository:** [Sentence Transformers on GitHub](https://github.com/UKPLab/sentence-transformers) |
|
|
- **Hugging Face:** [Sentence Transformers on Hugging Face](https://huggingface.co/models?library=sentence-transformers) |
|
|
|
|
|
### Full Model Architecture |
|
|
|
|
|
``` |
|
|
SentenceTransformer( |
|
|
(0): Transformer({'max_seq_length': 64, 'do_lower_case': False}) with Transformer model: MPNetModel |
|
|
(1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True}) |
|
|
(2): Asym( |
|
|
(anchor-0): Dense({'in_features': 768, 'out_features': 1024, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'}) |
|
|
(positive-0): Dense({'in_features': 768, 'out_features': 1024, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'}) |
|
|
) |
|
|
) |
|
|
``` |
|
|
|
|
|
## Usage |
|
|
|
|
|
### Direct Usage (Sentence Transformers) |
|
|
|
|
|
First install the required packages: |
|
|
|
|
|
```bash |
|
|
pip install -U sentence-transformers |
|
|
``` |
|
|
|
|
|
Then you can load and use the model with the following code: |
|
|
|
|
|
```python |
|
|
import torch |
|
|
import numpy as np |
|
|
from tqdm.auto import tqdm |
|
|
from sentence_transformers import SentenceTransformer |
|
|
from sentence_transformers.util import batch_to_device, cos_sim |
|
|
|
|
|
# Load the model |
|
|
model = SentenceTransformer("TechWolf/JobBERT-v2") |
|
|
|
|
|
def encode_batch(jobbert_model, texts): |
|
|
features = jobbert_model.tokenize(texts) |
|
|
features = batch_to_device(features, jobbert_model.device) |
|
|
features["text_keys"] = ["anchor"] |
|
|
with torch.no_grad(): |
|
|
out_features = jobbert_model.forward(features) |
|
|
return out_features["sentence_embedding"].cpu().numpy() |
|
|
|
|
|
def encode(jobbert_model, texts, batch_size: int = 8): |
|
|
# Sort texts by length and keep track of original indices |
|
|
sorted_indices = np.argsort([len(text) for text in texts]) |
|
|
sorted_texts = [texts[i] for i in sorted_indices] |
|
|
|
|
|
embeddings = [] |
|
|
|
|
|
# Encode in batches |
|
|
for i in tqdm(range(0, len(sorted_texts), batch_size)): |
|
|
batch = sorted_texts[i:i+batch_size] |
|
|
embeddings.append(encode_batch(jobbert_model, batch)) |
|
|
|
|
|
# Concatenate embeddings and reorder to original indices |
|
|
sorted_embeddings = np.concatenate(embeddings) |
|
|
original_order = np.argsort(sorted_indices) |
|
|
return sorted_embeddings[original_order] |
|
|
|
|
|
# Example usage |
|
|
job_titles = [ |
|
|
'Software Engineer', |
|
|
'Senior Software Developer', |
|
|
'Product Manager', |
|
|
'Data Scientist' |
|
|
] |
|
|
|
|
|
# Get embeddings |
|
|
embeddings = encode(model, job_titles) |
|
|
|
|
|
# Calculate cosine similarity matrix |
|
|
similarities = cos_sim(embeddings, embeddings) |
|
|
print(similarities) |
|
|
``` |
|
|
|
|
|
The output will be a similarity matrix where each value represents the cosine similarity between two job titles: |
|
|
|
|
|
``` |
|
|
tensor([[1.0000, 0.8723, 0.4821, 0.5447], |
|
|
[0.8723, 1.0000, 0.4822, 0.5019], |
|
|
[0.4821, 0.4822, 1.0000, 0.4328], |
|
|
[0.5447, 0.5019, 0.4328, 1.0000]]) |
|
|
``` |
|
|
|
|
|
In this example: |
|
|
- The diagonal values are 1.0000 (perfect similarity with itself) |
|
|
- 'Software Engineer' and 'Senior Software Developer' have high similarity (0.8723) |
|
|
- 'Product Manager' and 'Data Scientist' show lower similarity with other roles |
|
|
- All values range between 0 and 1, where higher values indicate greater similarity |
|
|
|
|
|
### Example Use Cases |
|
|
|
|
|
1. **Job Title Matching**: Find similar job titles for standardization or matching |
|
|
2. **Job Search**: Match job seekers with relevant positions based on title similarity |
|
|
3. **HR Analytics**: Analyze job title patterns and similarities across organizations |
|
|
4. **Talent Management**: Identify similar roles for career development and succession planning |
|
|
|
|
|
## Training Details |
|
|
|
|
|
### Training Dataset |
|
|
|
|
|
#### generator |
|
|
- Dataset: 5.5M+ job title pairs |
|
|
- Format: Anchor job titles paired with related skills/requirements |
|
|
- Training objective: Learn semantic similarity between job titles and their associated skills |
|
|
- Loss: CachedMultipleNegativesRankingLoss with cosine similarity |
|
|
|
|
|
### Training Hyperparameters |
|
|
- Batch Size: 2048 |
|
|
- Learning Rate: 5e-05 |
|
|
- Epochs: 1 |
|
|
- FP16 Training: Enabled |
|
|
- Optimizer: AdamW |
|
|
|
|
|
### Framework Versions |
|
|
- Python: 3.9.19 |
|
|
- Sentence Transformers: 3.1.0 |
|
|
- Transformers: 4.44.2 |
|
|
- PyTorch: 2.4.1+cu118 |
|
|
- Accelerate: 0.34.2 |
|
|
- Datasets: 3.0.0 |
|
|
- Tokenizers: 0.19.1 |
|
|
|
|
|
## Citation |
|
|
|
|
|
### BibTeX |
|
|
|
|
|
### JobBERT-v2 paper |
|
|
Please cite this paper when using JobBERT-v2: |
|
|
```bibtex |
|
|
@article{01K47W55SG7ZRKFG431ESRXC35, |
|
|
abstract = {{Labor market analysis relies on extracting insights from job advertisements, which provide valuable yet unstructured information on job titles and corresponding skill requirements. While state-of-the-art methods for skill extraction achieve strong performance, they depend on large language models (LLMs), which are computationally expensive and slow. In this paper, we propose ConTeXT-match, a novel contrastive learning approach with token-level attention that is well-suited for the extreme multi-label classification task of skill classification. ConTeXT-match significantly improves skill extraction efficiency and performance, achieving state-of-the-art results with a lightweight bi-encoder model. To support robust evaluation, we introduce Skill-XL a new benchmark with exhaustive, sentence-level skill annotations that explicitly address the redundancy in the large label space. Finally, we present JobBERT V2, an improved job title normalization model that leverages extracted skills to produce high-quality job title representations. Experiments demonstrate that our models are efficient, accurate, and scalable, making them ideal for large-scale, real-time labor market analysis.}}, |
|
|
author = {{Decorte, Jens-Joris and Van Hautte, Jeroen and Develder, Chris and Demeester, Thomas}}, |
|
|
issn = {{2169-3536}}, |
|
|
journal = {{IEEE ACCESS}}, |
|
|
keywords = {{Taxonomy,Contrastive learning,Training,Annotations,Benchmark testing,Training data,Large language models,Computational efficiency,Accuracy,Terminology,Labor market analysis,text encoders,skill extraction,job title normalization}}, |
|
|
language = {{eng}}, |
|
|
pages = {{133596--133608}}, |
|
|
title = {{Efficient text encoders for labor market analysis}}, |
|
|
url = {{http://doi.org/10.1109/ACCESS.2025.3589147}}, |
|
|
volume = {{13}}, |
|
|
year = {{2025}}, |
|
|
} |
|
|
``` |
|
|
|
|
|
#### Sentence Transformers |
|
|
```bibtex |
|
|
@inproceedings{reimers-2019-sentence-bert, |
|
|
title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks", |
|
|
author = "Reimers, Nils and Gurevych, Iryna", |
|
|
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing", |
|
|
month = "11", |
|
|
year = "2019", |
|
|
publisher = "Association for Computational Linguistics", |
|
|
url = "https://arxiv.org/abs/1908.10084", |
|
|
} |
|
|
``` |
|
|
|
|
|
#### CachedMultipleNegativesRankingLoss |
|
|
```bibtex |
|
|
@misc{gao2021scaling, |
|
|
title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup}, |
|
|
author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan}, |
|
|
year={2021}, |
|
|
eprint={2101.06983}, |
|
|
archivePrefix={arXiv}, |
|
|
primaryClass={cs.LG} |
|
|
} |
|
|
``` |
