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

Translate

Thursday, 27 September 2007

MATCH MAKER


Shalaka - Punam – Alex

This could have some relevance to MATCH MAKER (Which tool we have plaued to develop).

Some day, thousands of jobseekers would have downloaded "Match Maker" and hundred of these jobsskers may be"online" at the same time (on different job portals) search & "matching" the job - advts against their won resumes.We would have captured the "resumes" when they try to "Activate" this too we may also "capture" the job - advt being searched / matched by each jobseeker (using Matchmaker tool) while he is "Online" so ead user because our human job - advt "SPIDER ! Thye can "SERVE" matching job - advts to each user, as so as he comes online !

Hemen Parekh

Wednesday, 12 September 2007

SUBMIT RESUME FRESH GRADUATES

 

SN.

Website

email ID/chat

Jobsearch page URL

Search criteria

Log In

.

12

devseeker.in

http://www.devseeker.in/content/search.html

$\checkmark$ Keyword box

NO

Can be Incorporated


DESIGNS FOR "SUBMIT RESUME" FORMS - INDIA RECRUITER

Class of Candidates

IT

Non-IT

Fresh Graduates

Factory Workers

Clerical Staff

Main distinguishing factor/s

S/W or H/W Skills

Functions

Main Discipline/Branch

Trade

Vocation

Currently $\rightarrow$ 86 Skills

Currently $\rightarrow$ 43 Function

Principal "subjects"/ areas of study $\rightarrow$ B.A./B.Sc. etc $\rightarrow$ Politics/Chemistry

(as evidenced from designation) eg: Carpenter Welder Electrician

(as evidenced by designation) eg: Draftsman Receptionist Clerk (typist)

Can Main factor be used to generate "Population Frequency Distribution" Graph?

yes

yes

May be.

Unlikely

Unlikely

Graph will be plotted for? (Parameter for Xc axis)

Raw Score / Percentile Vs. % of Population

Raw Score / Percentile Vs. % of Population

% marks obtained in $\rightarrow$ final year $\rightarrow$ Average for all yrs

?

?

"Raw score" will be obtained from

Weightages of "Keywords" in each 86 Skills

Weightages of Keywords in each 43 Function

Can list of Main Factor grow with time?

yes

yes

yes

yes

yes.

What will be main source for expansion of Main factors?

Current Poll-&-Tricks

Current Poll-&-Tricks

Input U/I to be developed

Input U/I to be developed

Input U/I to be developed

User Generated New "Functions/Skills"

$\rightarrow$ Will these get transferred to "Submit Resume" form automatically?

NO

NO

NO

NO

NO

$\rightarrow$ If not, who will transfer & when?

We will transfer when we are satisfied that $\rightarrow$ These are broad/generic/universal, cutting across INDUSTRIES & DESIGNATIONS $\rightarrow$ Keywords for each are relevant/appropriate $\rightarrow$ At least each keyword has been voted for a minimum of 500 times

 

Class

IT

Non-IT

Fresh Graduates

Factory Workers

Clerical Staff

If the keywords under any given "Function/Skill" are only generic and not linked/related to any particular INDUSTRY, then we are in a ridiculous situation of developing a Common graph for, let us say "SALES" (Function) for professionals belonging to $\rightarrow$ Automotive Ind $\rightarrow$ Pharma Ind $\rightarrow$ Fin Service Ind $\rightarrow$ S/W Ind which is not a "Apple vs Apple Comparison" & is totally meaningless to most HR mgnt.

How do we solve this problem? How can we present to HR Mgnt. GRAPHS for people belonging to $\rightarrow$ same Function $\rightarrow$ same Design. Level

 

Message Batch

Resume Rater V 1.0

RR V 2.0

Undeclared?

1

04/08/07

21/09/07

14/08/07

2

14/08/07

26/09/07

3

24/08/07

4

03/09/07

5

07/09/07

13

14

14/10/07

16

16/11/07

Other Target Groups

  • Jobseekers
    • Starting Batch of 2.46 L $\quad \rightarrow$ Merge of Batch of 20,000 each
    • harvested thru. Resume Rater
  • Corporates
    • Starting batch of 40000/50000 $\rightarrow$ Merge of batch of 5000 each
    • Myriad batches $\quad$ 50000 +
  • Newspaper (Ed/Hrms etc) $\rightarrow$ Single batch $\quad$ 4"/1500
  • NHRD Network $\rightarrow$ -
  • Jobsites (from Jobbot directory) $\rightarrow$ 750 $\rightarrow$ on our website (Single batch)

Comments:

  • For each Target Group "Create such an EXCEL sheet to record details of
  • what message was sent $\rightarrow$ to whom (which batch) $\rightarrow$ To ensure that we do NOT duplicate same
  • when (date) message to same batch again $\&$ again
  • Keep a WEEK's gap between 2 messages $\qquad$ 13/09/07

to same BATCH

 






Friday, 7 September 2007

FUNCTIONAL SPECIFICATION FOR MATCH MAKER


Rahul

    Background

Millions of Jobseekers currently "Wast billions of manhours on various job sites, searching for "Suitable/ Matching" jobs - jobs which match their own skills / expertise / experience education.

Jobseeks try to narrow - down their serches by applying a no. of "Search criteria". Even then, most searches end - up in hundreds of search - results requiring many hours to open/ rea each job - advt to find out which ones are "Really" suitable/ matchin.

This happens due to:-

      . Large job - advt. database on website

      . Match - making algorithm, takes into account only the "search - parameters selected by that candidate

      . No attempt (no possibility) to match that candidates "RESUME" (skill keywords) with "Job Advt" (Skill Keywords).


When a candidate opens/ reads any job advt, he is, sub - consciously doing this matching. He knows this own skills. His own skill - related keywords are already "stored" in his memory when he read a job - advt, his brain is  picking - up the keywords contained in the job - advt and match compare these (keywords) with his own "Skill keywords".

If he finds many "matches",his brain tells him that this is a GOOD / MATCHING / RELEVANT job advt and worth "applying" for.

If he finds very few "matches" then it is a POOR match & not worth considering.

It is this "keywords" match making process that takes a big time since the brian must "read and compare". This process is made "works if there are hundreds of job - advt to be "read and matched".After reading/ matching a few, brain gets tired - even slackens. Match - making process deteriorates. Quality of decisions suffer.

the "relevance" (Match Index) established by the tool.

Enable candidate to "Apply online" against any GOOD / MATCHING job advt (shown in TABULATION) - even if on that particular job. portal.

To do this, our TOOL will need to be "pre - loaded" with the candidate's full text resume, once and for all (of course, he can EDIT anytime).

But the idea is that the RESUME WILL "reside" inside the TOOL and can be carried from one job - portal to another    job portal. A candidate need not submit his resume to dozens of job - portals !

Candidates will simply LOVE such a feature !

Candidate will be required to "preload" the MATCH MAKER with this text resume first, in order to connect to IndiaRecruiter - so that the TOOL can be "Activated". The MatchMaker itself can be downloaded from any no. of  website like Download.com

No need to assign User ID/ Password but this data (stored on our server will help us to find out, "how many of  those jobseekers who downloaded Matchmaker, came back to properly submit Resume on our website ?"
Unless duly/ properly "activated", tool cannot be used.

Tool will work on feature similar to MYRIAD - however MYRIAD's negative point is selection of

      First Link         Last Link

Which is combersome ! It would be nice if we can, somehow, improve upon this feature.

Either, all job - advt. Links get pickes up automatically OR may be Cut / Paste.

Job Advts which do NOT contain advertiser's email ID will be ignored rejected.

Because

Without advertiser's email ID our TOOL cannot forward the candidate's resume.

Point for Discussion:-

When a candidate clicks "Apply" he is ONLINE (may be on any job portal)

That might enable us to pick - up store in a database.

All the job - advts which a Candidate ever applied for, chronologically
[Database of Intentions = Job Apply History]

Trend analysis will tell us what kind of jobs, he is looking for !

This may be better than "harvesing" millons of job - advts!

Hemen Parekh