README.md

---
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}
}
```