Embedding

ENDLESS POSSIBILITIES AT YOUR FINGERTIPS

The ‘Profile & Job Embedding’ model analyzes the output of ‘Parsing’ and ‘Revealing’ layers and returns numerical vectors that represent a profile or job given as an input in a 1024-dimensional space.

The vector representation is computed by using the same HrFlow.ai technology used for ‘Scoring’ and ‘Reasoning’ layers. The vectors of similar profiles or jobs will be close to each other in the 1024-dimensional space. The ‘Profile & Job Embedding’ model can be used for organizing to unleash endless uses cases.

Now on, your AI experts Developers can focus on building great models instead of spending 90% of their time on pre-processing and vectorization.

Embedding (Vectors)

Dimension

Profile2vec

1024

Experience2vec

1024

Education2vec

1024

Skills2vec

1024

Job2vec

1024

Our Embedding API offers several different encoding levels, particularly when it comes to profiles, it can be a Profile Encoder Profile2vec or an Encoder Section (Experience2vec, Education2vec, Skills2vec).

Example of Profile Encoder "Profile2vec" (grouped by the profile job category)

Example of Section Encoder "Experience2vec" (grouped by the profile job category)

Why you should choose our Embedding ?

Features Workflow

Embedding Workflow
  • Tokenization : Our method of tokenization gives our models additional advantages in terms of making them less prone or prone to typing errors by leveraging information on subwords and multigrams.

  • Vectorization : Our vectorization method consists of a hierarchy of levels, a first one at the word level by using our in-house pre-trained word embeddings on the world largest HR entities dataset, then a second one for encoding paragraphs (sections in documents) based on Natural Language Processing state of the art models.

HrFlow.ai Vs Alternatives

Document embedding is the operation that consists of representing a document by a dense fixed length vector.

Embedding

Bag-of-words

Topic modelling

n-gram embeddings

Encoder-decoder

HrFlow

Multilingual

x

x

x

Can be trained in a cross-lingual approach

Use multilingual Word Embeddings

Sub-word information

x

x

x

Depends on the world-level embedding used

yes

HR Context Awareness

x

x

x

x

yes