Hi Friends,

Even as I launch this today ( my 80th Birthday ), I realize that there is yet so much to say and do. There is just no time to look back, no time to wonder,"Will anyone read these pages?"

With regards,
Hemen Parekh
27 June 2013

Now as I approach my 90th birthday ( 27 June 2023 ) , I invite you to visit my Digital Avatar ( www.hemenparekh.ai ) – and continue chatting with me , even when I am no more here physically

Friday, 19 June 2015

PLENTY OF POSSIBILITIES

ravi.apte@leftrightmind.com

gkapte@gmail.com

shuklendu.baji@sentientsystems.net



Dear Team,

The following abstract was copied from Royal Society site . Original news report appeared in MINT ( 17 June 2015 ) and titled :

" Your mobile phone data can reveal that you are unemployed " - Leslie D'Monte

In our mobile app " RESUME BLASTER " , Sentient System ( who developed it ) , have built a feature whereby , every time a User logs in , entire CONTACT DATA on that mobile, got picked up and got stored on our server !

Idea was to develop a " Pa-Rank " ( described in my email of Feb 2013 , below ) - provided a large number of jobseekers were to download and start using this App

As usual , I never got around ( without spending any money ! ) on promoting this app

But that possibility still persists  !

regards,

hcp

----------------------------------------------------------------------------------------

Tracking employment shocks using mobile phone data

( source : Journal of the Royal Society Interface / May 07 , 2015 )

Jameson L. Toole, Yu-Ru Lin, Erich Muehlegger, Daniel Shoag, Marta C. González, David Lazer


Can data from mobile phones be used to observe economic shocks and their consequences at multiple scales?

Here we present novel methods to detect mass layoffs, identify individuals affected by them and predict changes in aggregate unemployment rates using call detail records (CDRs) from mobile phones.

Using the closure of a large manufacturing plant as a case study, we first describe a structural break model to correctly detect the date of a mass layoff and estimate its size.

We then use a Bayesian classification model to identify affected individuals by observing changes in calling behaviour following the plant's closure.

For these affected individuals, we observe significant declines in social behaviour and mobility following job loss.

Using the features identified at the micro level, we show that the same changes in these calling behaviours, aggregated at the regional level, can improve forecasts of macro unemployment rates.

These methods and results highlight promise of new data resources to measure microeconomic behaviour and improve estimates of critical economic indicators.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

My Email dt Feb 2013

------------------------------

Dear  Team


Take a look at the following news report

That made me search in Google – and in all cases, found topmost entry at Wikipedia ( crowd sourcing , at its best ? ), the following :

Ø The Brave New World

Ø Karmarkar’s Algorithm

Ø Travelling Salesman Problem

Ø Swarm Intelligence

Ø Neural Network……….etc


In the article below , the company ( Relationship Science ),

Ø Does not depend upon “ user-generated content“,  …….but

Ø Scrapes the web for building its database

In either case , as Scott McNally ( Sun Micro ) famously said,

“ In internet age, privacy is dead. Period “

In our next app ( Resume Blaster ), Shuklendu has incorporated a Cookie which will pick-up the entire “ Contact Data “ from the smart phone of each user

Hopefully , someday in a later , non-commercial version of “ My Jobs “ , Prof Rahul Deshmukh will incorporate a similar cookie

And I suppose , in still later versions of both , we will succeed in picking-up the entire smart phone logs of voice-mails and text messages

I have absolutely no doubt that many apps are already doing this – so , if we ever get around to doing it , we will be merely “ re-inventing the wheel “ !

That will leave someone to write an algorithm that determines which “ Neural Connections “ are Strong / Average / Weak ,

based on the frequency and duration of the messages exchanged between them

Then , we would have beaten Goldman and his company, Relationship Science !

I may not be around to see this but I hope the idea survives

------------------------------------------------------------------------------------------------------------------

Google calls it PageRank – to decide the importance of a web page – then use this number to decide the rank ( position ) of that page within the search results

Let us follow Google ( very safe ! )

And call our “ Ranking Mechanism “ of any given resume , a “ Pa-Rank

Then arrange all arriving resumes ( in the mailbox of a Recruiter or on our web site ), in the descending order of their Pa-Rank

Highest ranked at the top

So , the Recruiter need to open/read only the top 5 / 7

In much the same manner that 95% of the searchers never go beyond the first 3 pages of any search-result despite millions of pages

Just imagine how much time it will save the recruiters to zero-in on the best candidates from amongst hundreds applying for any given job

This has the potential to revolutionize the On-line recruitment industry !

How can we determine a candidate’s Pa-Rank ?

Just like PageRank , which takes into account,

Ø No of Incoming links to a given page

Ø Importance of each incoming link ( based on its own incoming links )

No doubt , Google’s actual algorithm must be very complex – and continues to evolve

Thru Resume Blaster ( and someday , thru My Jobs as well ), we pick-up the “ Contact List “ ( of mobile nos ) of say, Ajay ( a user registered on our site )

These are his “ Primary Incoming Links “ – say , 40

Out of these 40 , say , 10 have also registered on our site AND , are using Resume Blaster

In turn , these 10 have , between them , 600 contacts ( “ Secondary Incoming Links “ )

So , for Ajay , total Incoming links = 40+600 = 640

Now , if Ajay thinks he has figured out how our software computes Pa-Rank then he will somehow find out the mobile number of Mukesh Ambani , and add into his “ Contacts “ – assuming that at one stroke ,he can add ONE MILLION incoming links to himself !

But , we know that such high-flying executives are unlikely to have registered on our site – much less , to be using Resume Blaster !

Hence we treat this as a “ Dead Incoming Link

On top of that , we can further refine our Pa-Rank based on,

Ø The frequency with which Ajay keeps calling each of those 10 persons ( who have registered on our site AND are also using Resume Blaster )

Ø The duration of each call ( or even text message ? )

We can multiply / sum-up / find average etc of such “ exchanges of messages – both ways “ and arrive at some “ FACTOR “

Then multiply the Total No of Live Incoming Links with this factor , to compute PaRank

Looks crude , no doubt !

But Google has proved that it works !

What PageRank is for all the web pages in the World , someday , PaRank will be to all the resumes ( which too , like any web page , are documents ) , in India

We must do it !

On internet , “ Ideas are dime-a-dozen “

But elegant executions are rare !

Regards

hemen

No comments:

Post a Comment