How does the AI behind the "People you may know" work?
I'm amazed by its accuracy. Just today, it suggested over a dozen people that I do know and am connected to in one way or another. What's the logic behind the code, in simple terms? Is there some code written in the contacts file we upload? Although, some of the people that I "know" weren't hidden somewhere in my address book, so it can't be that. Is there some geo localization at work, or is it a sort of datamining AI? I'm so curious to know how it works!
Clarification added June 24, 2007:
It can't be from my "other" contacts because I don't have any stored there. Nor can it be keywords that we have stored in our profiles because some of the "people i may know" have completely different backgrounds, education, jobs, etc.
Clarification added June 24, 2007:
I think there is a cross co-relation between contacts that I upload and those uploaded by others that include me. Even if we don't invite these contacts or even if these contacts don't invite me, then subsequently delete them from our contact list, I suspect LinkedIn stores this information.
The weirdest thing though is when it suggests someone I only know "off the street": someone who's contact info I don't have and vice versa and someone with whom I have no history with (school, employer, same industry, etc.).
Good Answers (1)
David N
Technologist, Evangelist, Entrepreneur, Inventor ► Executive Consultant, CTO
Best Answers in: Using LinkedIn (229), Software Development (21), Computers and Software (20), Web Development (17), Customer Service (10), Staffing and Recruiting (10), Career Management (10), Small Business (7), Enterprise Software (7), Project Management (6), Quality Management and Standards (6), Starting Up (6), Accounting (5), Business Development (5), Organizational Development (5), E-Commerce (5), Computer Networking (5), Wireless (5), Business Analytics (4), Professional Networking (4), Certification and Licenses (3), Internationalization and Localization (3), Internet Marketing (3), Public Relations (3), Customer Relationship Management (3), Distribution (3), Market Research and Definition (3), Ethics (3), Telecommunications (3), Purchasing (2), Job Search (2), Mentoring (2), Mergers and Acquisitions (2), Compensation and Benefits (2), Corporate Law (2), Intellectual Property (2), Advertising (2), Writing and Editing (2), Corporate Governance (2), Planning (2), Non-profit Management (2), Manufacturing (2), Product Design (2), Franchising (2), Blogging (2), Regulation and Compliance (1), Travel Tools (1), Education and Schools (1), Freelancing and Contracting (1), Conference Venues (1), Budgeting (1), Venture Capital and Private Equity (1), Economics (1), Financial Regulation (1), Risk Management (1), Personnel Policies (1), Exporting/Importing (1), Treaties, Agreements and Organizations (1), Finance and Securities Law (1), Employment and Labor Law (1), Direct Marketing (1), Events Marketing (1), Graphic Design (1), Change Management (1), Labor Relations (1), Equity Markets (1), Inventory Management (1), Personal Taxes (1), Branding (1), Positioning (1), Biotech (1), Databases (1), Information Security (1), Information Storage (1)
I believe LinkedIn starts by looking for LinkedIn members whom you've imported but not connected (Other Contacts), who are local to your area, and also looks for 2nd degree connections.
More Answers (12)
Sheilah E
Owner, ★SME Management:.......... Business Management and Accounting Consultant
Best Answers in: Using LinkedIn (884), Professional Networking (39), Staffing and Recruiting (30), Accounting (20), Government Policy (18), Job Search (17), Career Management (15), Ethics (15), Business Development (14), Customer Service (12), Mentoring (12), Education and Schools (10), Computers and Software (10), Property Law (9), Organizational Development (9), Small Business (9), Personnel Policies (8), Criminal Law (8), Advertising (8), Internet Marketing (8), Labor Relations (7), Non-profit Management (7), Starting Up (6), Blogging (6), Purchasing (5), Government Services (5), Compensation and Benefits (5), Tax Law (5), Lead Generation (5), Planning (5), Manufacturing (5), Quality Management and Standards (5), Web Development (5), Corporate Law (4), Direct Marketing (4), Writing and Editing (4), Corporate Governance (4), Change Management (4), Communication and Public Speaking (4), Professional Organizations (4), Software Development (4), Facilities Management (3), Regulation and Compliance (3), Travel Tools (3), Freelancing and Contracting (3), Auditing (3), Venture Capital and Private Equity (3), Economics (3), International Law (3), Internationalization and Localization (3), Treaties, Agreements and Organizations (3), Market Research and Definition (3), Business Plans (3), Information Security (3), Business Dining and Entertainment (2), Resume Writing (2), Government Contracts (2), Employment and Labor Law (2), Customer Relationship Management (2), Sales Techniques (2), Business Analytics (2), Derivatives Markets (2), Inventory Management (2), Project Management (2), Supply Chain Management (2), Individual Insurance (2), Personal Taxes (2), Personal Real Estate (2), Product Design (2), Pricing (2), Incorporation (2), E-Commerce (2), Enterprise Software (2), Computer Networking (2), Telecommunications (2), Wireless (2), Air Travel (1), Certification and Licenses (1), Occupational Training (1), Conference Planning (1), Budgeting (1), Corporate Debt (1), Financial Regulation (1), Risk Management (1), Exporting/Importing (1), Offshoring and Outsourcing (1), Customs, Tariffs and Taxes (1), Contracts (1), Finance and Securities Law (1), Viral Marketing (1), Graphic Design (1), Public Relations (1), Hedge Funds (1), Non-profit Fundraising (1), Philanthropy (1), Personal Debt Management (1), Retirement and Estate Planning (1), Wealth Management (1), Branding (1), Positioning (1), Energy and Development (1), Biotech (1), Information Storage (1)
I just asked one of the IT people from linkedin about this. His response was that the people in these lists have some of the same keywords in their profile that we have in ours. It may be a job description or a similar name for a business etc.
Sheilah
Obviously it is multifactor. Mine shows a couple people with whom I've only emailed a couple times (contact), another I worked with (he as a vendor)(keyword), and a couple I have no idea where they came from since I don't know them. The vendor entry amazed me as he had left the company some time ago.
Very interesting, people that I actually know keep showing up ... I suspect neiboring zip codes are a factor in the algorithm.
Clarification added June 24, 2007:
I meant "neighboring" ,,, sorry about the typo
I'm seeing a lot of "off the street" people I know being suggested as well. It's more than weird, although I wouldn't use the word creepy just yet. When I logged in just now the five suggestions were all people that I've sent mail via GMail to, socially not professionally. And similarly, there is no other common thread except for that-- not previous school, employer, etc. I've never imported contacts from GMail.
The keyword / geolocation answer doesn't hold water for me. I'd like to know how it works, and whether it can be disabled.
This feature's added people who I haven't got in my Outlook contacts and who aren't related to me in any way other than one or two emails sent by via a webmail client to a secondary address.
The people I know are stored contacts in my Hotmail account...
Even people I have one mailed with years ago.
LinkedIn seems to read those...
Don't use outlook etc
Steven S
Research Scientist and Sr. Product Manager
Best Answers in: Using LinkedIn (44), Education and Schools (2), Regulation and Compliance (1), Corporate Taxes (1), Internationalization and Localization (1), Organizational Development (1), Web Development (1)
"People you may know" is powered by a sophisticated predictive model that uses many factors to guess people you might know.
It's still in beta, and we've made some significant refinements to it recently.
It's pretty cool, no? Please give us feedback on it.
I'm amazed!!! One penny for a glance at that "predictive algorithm"?
People, I havent seen for years or even thought of is appearing on that list.
Though one hint, a few years ago I uploaded my email adresses from Outlook - could it be that you saved these addresses from me?
Thx
Steffen Vogdrup
I was just given two names of people that I do know, but that I have no connection with through LinkedIn. My profile is not filled in with enough data OR the right data to make a connection. Therefore, how would they know I worked with them, if that employer is not listed, how would they know I went to school with them, when that inststution is not listed. I am not a regular LinkedIn user and only have a few connections, and those connections have no connections to the people on my list. SO.... where did those names come from?
Jennifer J
CISSP, Network Security Specialist & Industry Blogger
Best Answers in: Quality Management and Standards (1), Computer Networking (1), Information Security (1)
I'm equally amazed. It found someone in a different industry, but that I do know, not connected at any degree, but he was connected through the same Internet provider... could it pull the providers IP and factor that in? Who knows...
My personal belief is the entire thing is a consipiracy, lead by the MAGIC 8 BALL....
That's what I think...
-jj
Links:
Clarification added June 27, 2007:
Oooh I have more! (The MAGIC 8 BALL has spoken again)
There are 2 people in my PYMK (Peeps you may know)
1) Person A is already in my network, a direct (1st) contact
2) Person B is not in my contact, nor anyone else's contact.
- Person B has an account, but has 0 contacts.
- Person B is several states away SouthWest US
- Person B is not in the same industry
- I do not know Person B, nor do I have any other contacts at his company
... the deductive reasoning continues
Dan L
Innovative, creative and dependable leader with a track record of accomplishments in marketing, technology and strategy.
Best Answers in: Staffing and Recruiting (1), Internet Marketing (1), Viral Marketing (1), Customer Relationship Management (1), Business Analytics (1), Career Management (1), Professional Networking (1), Business Plans (1), E-Commerce (1), Databases (1)
It is very cool. For me, it found:
* someone who moved on from a school I attended;
* Someone who I knew as a vendor to the company I worked for.
* several folks close in years when I completed my undergrad degree.
It has not yet found someone for me "off the street". I have no contacts listed, and LinkedIn does not know about my Outlook contact list.
David B
Software Architect - Lead Software Engineer - Scrum Master at Partezis
Best Answers in: Software Development (29), Web Development (8), Computers and Software (5), Databases (3), Career Management (2), Education and Schools (1), Business Analytics (1), Project Management (1), Professional Books and Resources (1), Biotech (1), Blogging (1), Enterprise Software (1), Information Security (1), Information Storage (1)
There isn't any kind of magic here, and even less the slightest trace of artificial intelligence. This is simply based on graph theory, starting exploration of the graph of your relations from your node, and filtering those relations according to what is called "homophily". The greater is the homophily between two nodes, the more likely two nodes will be connected. For a good introduction on the social network theory, see the document linked below.
