Embedding API
Build powerful machine learning apps with pre-trained HR field vector representation.
Save 90% of research time, build powerful model with small data and avoid representation biases.
Join 1,000+ businesses on HrFlow.ai
Focus on building quickly highly
accurate models, not features engineering
The Embedding API analyzes the output of the Parsing API. and Revealing API. to return numerical vectors that represent a profile, a job or an HR document section given as an input in a 512-dimensional space.
The vector representation is computed by using the same HrFlow.ai technology used for the Searching API, Scoring API and Reasoning API.
The vectors of similar objects will be close to each other in the 512-dimensional space. These vectors can be used by organizations to unleash endless uses cases. Now on, your AI experts and Developers can focus on building great models instead of spending 90% of their time collecting large datasets, preprocessing data and features engineering.
Broad use cases
Similarity analysis, Search and retrieval, Machine transfer learning.
Multiscale modeling
Large portfolio of vector families to satisfy all your use cases: Profile2Vec, Skills2Vec, Job2Vec, Interests2Vec, Education2Vec, Experience2Vec.
Representation debiasing
Fairness is not a default. That‘s why we built inclusive models that measures and mitigates unintended bias in text.
N-gram information
Combinations of adjacent words that have meaning together.
Sub-word information
Enriched word vectors with relevant sub-information.
Multiple Languages
Supports 32+ languages.
6
Vector families
Statisfy all your use cases
32+
Languages
Scale up globally
90%
Time saving
Build models fast
Trusted by HCM companies and forward-thinking HR leaders
Explore what our clients say
10,000+
predictions per day
photo
Olivier PONTECAILLE
« At Fed Group, structured candidate data is at the core of our business strategy to deliver an exceptional and personalized customer service. HrFlow.ai helped us to automate data classification at scale avoiding manual entry.. »
Works with the tools you use
We lift the load on your research team,
so they can focus on building great models
6 years of expertise in representing HR documents
Embedding
TF-IDF
L.D.A
Doc2Vec
Seq2Seq
Transformer
HrFlow.ai
N-grams
Multilingual
Sub-words
Dimensionality
Reduction
Neural
Memory
Semantic
Compression
Quantization
Unicode
Allocation
bias
Representation
bias
Multi scale
modeling
Allocation
bias
Hierarchical
representation
HR Data

Getting started is easy