ONA - organizational network analysis - is becoming an important topic for HR-technology. Simply put, ONA provides insight into how organizations really function.
Embedding ONA capability has the potential to enable employers and employees to organize themselves more effectively, communicate more impactfully, and to lead their companies forward.
9711147426✨Call In girls Gurgaon Sector 31. SCO 25 escort service
ONA ( organizational network analysis ) to enable individuals to impact their organization - part 2
1. Facebook social
widget
For content discovery
Form of social pressure (2 friends like this)
Two of your close friends like this
- You are more likely to
discover the content
- You are more inclided to
believe/like the content
4. Create a blueprint for Future of Work in Factories. Test the
prototype in the customer production Factory, and obtain
first feedback of the prototype from end-users.
Similar Contributions:
Future of Onboarding Learning on Shopfloor
Alicia
Young
Bob
Stone
Joe
McNamara
Yves
Delacroix
Learning:
Learn to make
software Prototypes
Prototyping
Future of Work Factories of the Future
Experimental
Design
Text is analyzed in real time
- Extract Topic Meaning
- Keyword Extraction
Keywords are displayed
Key Influencers are found. You can:
- See their contribution
- Contact them directly
Related Marketplace Projects
Relevant Learning is found.
Add to MyLearning
Create a «Future of Work in Factories» Concept
5. Similar Contributions:
Future of Onboarding Learning on Shopfloor
Alicia
Young
Bob
Stone
Joe
McNamara
Yves
Delacroix
Future of Work Factories of the Future
Contact Alicia directly through the systemCheck Alicia´s contributions
- See if you shoud align
Align & Connect
Align with people working on the similar things
Find people from my network (that are in my communication
neighborhood)
Find key influencers company-wide
Circle Indicates people from my immediate
network
6. New Marketplace Project
Title
Description
Working on similar topics:
Bob
Stone
Joe
McNamara
Self-driving buses
The future of public transportation is in self-driving buses, that
are driving through metropolitan and rural areas.
The goal of this project is to create software for self-driving
buses, which would be the backbone of the public
transportation in metropolitan areas.
Self-driving Public Transport Buses
Interested in Contributing:
Yves
Delacroix
Text is analyzed in real time
- Extract Meaning from text
- Keyword Extraction
- Keyword translation to users language
ONA network finds key people to contact
- Based on keywords
- Based on Impact/Influence
Yves is interested in the topic
- Shown as suggested for the project
Invite Yves to contribute
7. Our Approach: Privacy and GDPR
Easier than feared!
Privacy-by-design (GDPR):
a) Article 4(5) defines pseudonymization as “the
processing of personal data in such a way that the
data can no longer be attributed to a specific data
subject without the use of additional information.”
b) For summary statistics or groups (non-PII) analytics:
Recital 26 defines anonymized data as “data
rendered anonymous in such a way that the data
subject is not or no longer identifiable.”
Cleanup PII
Utilize
PII re-bind
ONA/NLP/ML
Pseudonymize
Dataset
User
Consent
(PII Separation)
NOTE: PII - Personally identifiable information
7
8. What next?
8
1. Roll-out a GDPR-aware MVP
2. Learn from it and grow Data Product Managers
3. Apply ONA metrics & ML to pioneer new products
9. NLP and Keyword extraction
Input: Training email dataset
• Sanitization, header removal
• Filter outbound data
Processing:
• Simple NL preprocessing (lowercase, remove special
characters, tokenize)
• Named entity extraction and removal
• Stopwords removal
11. .
LDA and RAKE
LDA – Latent Dirichlet Allocation:
• Topic modeling algorithm
• Extracts relevant topics from a corpus
• Unsupervised learning
• Needs number of topics beforehand
RAKE – Rapid Automatic Keyword Extraction:
• Statistical modeling of text
• Creates rank of keywords
• Easy to run, not actually ML
12. .
Word embeddings and TF/IDF
Word embeddings:
• Used fasttext embeddings – multilanguage
• Expand found keywords by similarity vectors
• Threshold value for similarity
TF/IDF:
• Ranking and relevance for documents
• Using any Lucene based solution
• Get free n-grams and typos
13. More Info?
Visit our Article Series:
bit.ly/2Rwhg2R
13
Vladimir Mijatovic | Alexandru Filip | Agron Fazliu
February, 2019
Freiburg