Talent Sourcing and Matching 
Artificial Intelligence & 
Black Box Semantic Search 
vs. 
Human Cognition & Sourcing 
Glen Cathey 
www.linkedin.com/in/glencathey 
www.booleanblackbelt.com
What’s the big deal anyway? 
Some people believe resume, LinkedIn and Internet 
sourcing is so easy that sourcing is either dying or 
dead or can be performed for $6/hour
Resume and LinkedIn sourcing 
appears simple and easy on the 
surface, however – it is deceptively 
difficult and complex
 Anyone can find candidates because all searches 
"work" as long as they are syntactically correct 
 That doesn’t mean the searches are finding all of the 
best candidates! 
 People make assumptions when creating 
searches 
 Every time an assumption is made, there is room for 
error and you unknowingly miss and/or eliminate 
results!
 No single search can return all potentially 
qualified people 
 Every search both includes some qualified 
people and excludes some qualified people 
 Some of the best people have resumes or 
social profiles that may not appear to be 
obvious or strong matches to your needs
 People cannot effectively be reduced to and 
represented by a text-based document 
 Job seekers are NOT professional resume or 
LinkedIn profile writers 
 Most people still believe shorter and concise 
resumes and social profiles are still better 
 This means they are removing data/info from their 
resumes which can no longer be searched for!
 No one mentions every skill or responsibility 
they’ve had, nor describes every environment 
they’ve ever worked in 
 There are many ways of expressing the same 
skills and experience 
 Employers often don’t use the same job titles 
for the same job functions
 People don’t create their resumes and 
LinkedIn profiles thinking about how you will 
search for them 
 Sometimes people don’t even use correct 
terminology 
 Anyone easy for you to find is easy for other 
recruiters to find = no competitive 
advantage!
In addition to the people you do find, there are 
Dark Matter results of people that exist to be 
retrieved, but can't be found through standard, 
direct or obvious methods 
I estimate Dark Matter to be at least 
50% of each source searched
Finding some people is 
easy…
Finding all of the best 
people IS NOT!
“When every business has free and ubiquitous 
data, the ability to understand it and extract 
value from it becomes the complimentary scarce 
factor. It leads to intelligence, and the intelligent 
business is the successful business, regardless of 
its size. Data is the sword of the 21st century, 
those who wield it well, the Samurai.” 
-Jonathan Rosenberg, SVP, Product Management @ Google
 Stop wasting time trying to create difficult and 
complex Boolean search strings 
 Let "intelligent search and match applications" 
do the work for you 
 A single query will give you the results you 
need - no more re-querying, no more waste of 
time!
 Understand titles, skills, and concepts 
 Automatically analyze and define relationships 
between words and concepts 
 Intuit and infer experience by context
 Perform pattern recognition 
 Employ semantic search 
 Perform fuzzy matching
How do they really work?
 Intuit experience by context = resume parsing 
 Parsing breaks down and extracts resume 
information 
 Most recent title and employer 
 Skills and experience 
 Years of experience – overall, in each position, with 
specific skills, in management, etc. 
 Education
 Parsing enables structured, fielded search 
 Search by: 
 Most recent title 
 Recent experience 
 Years of experience 
 Etc.
 Well developed ontologies and taxonomies 
 Hierarchical
 Synonymous terms 
 Programmer, Software Engineer, Developer 
 Tax Manager, Manager of Tax 
 CSR, Customer Service Representative 
 Ruby on Rails, RoR, Rails, Ruby 
 Oracle Financials, Oracle Applications, e-Business 
Suite, etc.
 Some applications use complex statistical 
methods in an attempt to "understand" 
language and the relationships between words 
 Example: Google Distance
 Keywords with the same or similar meanings in 
a natural language sense tend to be "close" in 
units of Google distance, while words with 
dissimilar meanings tend to be farther apart
 A measure of semantic interrelatedness 
derived from the number of hits returned by 
the Google search engine for a given set of 
keywords
 Non-interactive and unsupervised machine 
learning technique seeking to automatically 
analyze and define relationships between 
words and concepts 
 Clustering is a common technique for statistical 
data analysis
 The design and development of algorithms 
that allow computers to evolve behaviors 
based on empirical data 
 A major focus is to automatically learn to 
recognize complex patterns and make 
intelligent decisions and classifications based 
on data
 Aims to classify data (patterns) in resumes 
based either on a priori knowledge or on 
statistical information extracted from the 
patterns 
 A priori: independent of experience 
 Example of pattern recognition: spam filters
 Finds approximate matches to a pattern in a 
string 
 Useful for word and phrase variations and 
misspellings
 Reduce time to find relevant matches 
 Can lessen or eliminate the need for recruiters 
to have deep and specialized knowledge within 
an industry or skill set 
 Reduce and even eliminate time spent on 
research
 Go beyond literal, identical lexical matching 
 Levels the playing field 
 Can make an inexperienced person look like a 
sourcing wizard 
 Good for teams with low search/sourcing capability
 Work well for positions where titles effectively 
identify matches and where there is a low 
volume and variety of keywords 
 Good for a high volume of unchanging hiring 
needs
 Removes thought from the talent identification 
and decision making process 
 Danger of eliminating the need for recruiters to 
understand what they’re searching for 
 Information technology, healthcare, and other 
sectors/verticals can create pose serious 
challenges to matching apps
 Apps find some people, bury or eliminate 
others 
 Is finding some people good enough for your 
organization? 
 Shouldn’t your goal be to find ALL of the BEST 
people?
 Matching apps level the playing field 
 People from different companies using the same 
solution will both find and miss the same people 
 Competitors using the same search and match 
solution will have no competitive advantage over 
each other!
 Belief that one search finds all of the best 
candidates is intrinsically flawed and simply not 
based in reality 
 Top talent isn't represented by what a search 
engine "thinks" has the best resume or profile 
 AI and semantic search apps favor keyword rich 
resumes and profiles
 Keyword poor resumes and profiles may in fact 
represent better talent than keyword rich 
resumes and profiles 
 It’s not just a matter of keyword frequency or 
even keyword presence! 
 AI powered search & match applications can 
only return results that explicitly mention 
required keywords and their variants
 Many people have skills and experience that 
are simply not mentioned anywhere in their 
resumes! 
 These people are the Dark Matter of 
databases, ATS’s, and social networks, and 
they exist but cannot be found via direct 
search/match methods – AI or otherwise!
 Pre-built taxonomies are static, limited in their 
completeness and must be continually updated 
in order to stay relevant and effective 
 Taxonomies are only as good as who created 
them 
 Applications can only match on what’s present 
and cannot “think outside of the box”
 Semantic clustering and NLP applications can 
retrieve related search terms, but that does 
not mean they are relevant for your need!
 Match primarily on titles and skill terms 
 True match is at the level of role, responsibilities, 
environment, etc. 
 Some applications rank results favoring recent 
employment duration 
 Is someone who has been in their current company 
for 5 years really “better” than someone who has 
been with their current company for 2 years?
 Apps don’t "know" what you’re looking for or 
what's the best match for your company 
 Apps are not and cannot be "aware" of people 
that were excluded from their search results 
 Applications are not truly intelligent – they do 
not actually "know" or "understand" the 
meaning of titles and terms
 The ability to learn or understand or to deal 
with new or trying situations 
 The ability to apply knowledge to manipulate 
one’s environment or to think abstractly 
 REASON; the power of comprehending and 
inferring 
Source: Merriam-Webster.com
 The capability of a machine to imitate 
intelligent human behavior 
 Artificial = humanly contrived 
Source: Merriam-Webster.com
 Dr. Michio Kaku 
 Theoretical physicist and futurist 
specializing in string field theory 
 Harvard Grad (summa cum laude) 
 Berkeley Ph.D 
 Currently working on completing 
Einstein's dream of a unified field 
theory 
 What are his thoughts on AI?
 “…pattern recognition and common sense are 
the two most difficult, unsolved problems in 
artificial intelligence theory. Pattern 
recognition means the ability to see, hear, and 
to understand what you are seeing and 
understand what you are hearing. Common 
sense means your ability to make sense out of 
the world, which even children can perform.” 
- Dr. Michio Kaku
 Dr. Michio Kaku believes the job market of the 
future will be “dominated by jobs involving 
common sense (e.g. leadership, judgment, 
entertainment, art, analysis, creativity) and 
pattern recognition (e.g. vision and non-repetitive 
jobs). Jobs like brokers, tellers, 
agents, low level accountants and jobs 
involving inventory and repetition will be 
eliminated.”
 That’s good news for sourcers and recruiters who 
perform sourcing! 
 Sourcing requires judgment, creativity, analysis, 
common sense and pattern recognition (instantly 
making sense of human capital data) 
 Sourcers of the future will be human capital data 
analysts who are experts in HCDIR & A – Human 
Capital Data Information Retrieval and Analysis
 Matching apps do not have the dynamic ability 
to learn, understand and instantly relate new 
concepts and through direct experience and 
observation 
 They depend on taxonomies, statistical 
models, or semantic clustering to “understand” 
relationships and concepts
 The human mind naturally organizes its 
knowledge of the world, instantly relating new 
terms and concepts and judging their relevance
 Example: A sourcer who is completely 
unfamiliar with “infection control” can instantly 
recognize non-highlighted but related and 
relevant terms and incorporate them into new 
and improved searches 
Carolinas HealthCare System, Charlotte, NC 
Infection Preventionist 1997-present 
Responsible for all aspects of infection prevention and control for an 
800+bed hospital. Uses science-based research to perform infection 
prevention. Conducts all aspects of surveillance, data analysis, and 
presents data to interdisciplinary teams, including the Infection 
Control Committee.
 Human sourcers can learn from research and 
search results, dynamically and adaptively 
identifying related and relevant search terms 
and incorporate them into successive searches 
to continuously refine and improve searches 
for more relevant results
 For example, if a recruiter was sourcing for a 
position that required a skill that they were 
unfamiliar with (e.g.,“Cockburn Use Case 
Methodology” ) they could quickly perform 
research to learn more about it 
 In the next slide, you will see a screen capture 
of such research
 From this quick research , the recruiter would 
be able to determine that most people would 
not explicitly mention “Cockburn Use Case 
Methodology,” let alone “Cockburn” (which the 
research revealed is pronounced “Co-burn”) – 
thus they would not include the term in their 
searches
 Instead, it would be a better idea to search for 
candidates that mention experience with Agile 
methodology and simply call and ask them if 
they have experience with using Cockburn’s 
use case methodology (which many likely 
would)
 Applications using Natural Language Processing 
do not truly understand human language 
 They use complex statistical methods to resolve 
the many difficulties associated with making 
sense of human language 
 NLP experts admit that to computers, even simple 
sentences can be highly ambiguous when 
processed with realistic grammars, yielding 
thousands or millions of possible analyses
 Humans effortlessly and automatically process 
and understand language, regardless of 
sentence length or complexity, ambiguity, 
incorrect grammar, etc. 
 We can udnretsnad any msseed up stnecene as 
lnog as the lsat and frsit lteetrs of wdros are in 
the crrcoet plaecs
 Human sourcers and recruiters can deduce 
potential experience, even in the absence of 
information (not explicitly mentioned in the 
resume/profile) 
 Applications can only work with what’s 
actually mentioned in a resume – if it's not 
explicitly mentioned, it can't match on it
 Applications are not aware that many of the 
best people have average resumes 
 Applications are not aware of the people their 
algorithms bury in results or eliminate entirely 
 Human sourcers can become aware of and 
specifically target this Dark Matter
How can you target resumes and 
LinkedIn profiles that exist, but 
your searches can’t and don’t 
retrieve them?
 Well developed taxonomies, 
semantically generated query 
clouds and matching 
algorithms can help greatly 
with automatically searching 
for and matching on 
synonymous terms, related 
words, word variants, 
misspellings, etc.
Think + Perform Research 
 For keyword, phrase or title you are thinking 
of using in your search, realize: 
1. Not everyone will explicitly mention what you 
think they would or should mention in their 
resume/profile 
2. There are many different and often unexpected 
ways of expressing the same skills and experience
Global Experience 
What search terms might you use if you are 
looking for people with global experience? 
How many can you think of off the top of your 
head?
In a few minutes of exploratory research, a sourcer can 
come up with a volume of related and relevant terms 
 Global, international, foreign, multinational, 
worldwide 
 Europe, European, EU, EMEA, Asia, Asia-Pac, Pacific 
Rim, South America, Latin America, Americas, CALA 
(Caribbean and Latin America), Middle East 
 Canada, Japan, China, Russia, India, UK, United 
Kingdom, etc. 
 Countries, Offshore, Overseas
How can you target results of people 
that your searches retrieve but the 
results are buried (ranked poorly or 
"too many" results to be reviewed) and 
you don’t find them?
 Search and matching software powered by 
artificial intelligence / black box semantic 
search doesn't have a solution to this 
challenge 
 One of the major claims AI/semantic search 
applications make is that their solutions can 
find the "right people" in one search
 However - a single search strategy is 
intrinsically flawed and limited - no single 
search can find all qualified candidates, and 
each search both includes qualified people as 
well as excludes qualified people 
 I am not aware of any search & match 
software that allows for successive searching 
via mutually exclusive filtering
Run Multiple Searches 
 Start with maximum qualifications 
 Use the NOT operator to systematically filter 
through mutually exclusive result sets 
 End with minimum qualifications
 Required: A,B,C 
 Explicitly desired: D,E 
 Implicitly desired: F
1. A and B and C and D and E and F 
2. A and B and C and D and E and NOT F 
3. A and B and C and D and NOT E and F 
4. A and B and C and NOT D and E and F 
5. A and B and C and NOT D and NOT E and F 
6. A and B and C and D and NOT E and NOT F 
7. A and B and C and NOT D and E and NOT F 
8. A and B and C and NOT D and NOT E and NOT F
Search #1 
Search #8
Probability-Based and Exhaustive! 
This approach allows for: 
1. The specific targeting of people who theoretically have 
the highest probability of being a match based on 
information present 
2. The specific targeting of people who may be the best 
match, but may have keyword/information poor resumes 
or profiles, who do not explicitly mention what you think 
the "right" person would or should mention 
3. The ability to systematically filter through all available 
results via manageable and mutually exclusive result sets 
– never seeing the same person twice!
 A mix of “man and machine,” integrating 
human knowledge and expertise into computer 
systems 
 Essentially - the best of both worlds: 
 Autopilot: An artificially intelligent semantic 
matching engine 
 Manual Override: Ability to take complete control 
over searches and search results
 An artificial intelligence semantic matching 
engine coupled with taxonomies built by 
human SMEs that are continually modified and 
improved specifically for the organization 
 No COTS solution is customized for any specific 
employer, industry or discipline, nor 100% 
complete
 Resume and LinkedIn profile parsing 
 Structured, contextual search 
 Most recent title and experience, overall years of 
experience, education, etc. 
 White Box relevance weighting 
 Configurable by users – no black box! 
 Searchable tagging for level 5 semantic search
 Standard and extended Boolean in full text and 
field-based search 
 AND, OR, NOT, configurable proximity, weighting 
 Configurable proximity enables level 3 semantic 
search 
 Variable term weighting allows users to control 
which search terms are more important and thus 
control over true relevance
 Lucene is a free and open source text search 
engine that support configurable proximity and 
term weighting, and can be integrated into 
some existing ATS's/databases 
 Some Applicant Tracking Systems already have 
databases powered by text search engines that 
allow for extended Boolean
“Society has reached the point where 
one can push a button and immediately 
be deluged with…information. This is all 
very convenient, of course, but if one is 
not careful there is a danger of losing 
the ability to think.” 
- EijiToyoda
 Data and information requires analysis to 
support decision making 
 Just as very expensive Business Intelligence 
and Financial Analytics software hasn't 
replaced the need for people to make sense of 
the data, there is no software solution for HR 
and recruiting that replaces the need for 
people to analyze and interpret human capital 
data to make appropriate decisions
 Matching apps move/retrieve information, but 
only PEOPLE can analyze and interpret for 
relevance and make intelligent decisions 
 Relevant: the ability (as of an information retrieval 
system) to retrieve material that satisfies the 
needs of the user [1] 
 Only the user (sourcer/recruiter) can judge 
relevance! 
[1] Source: Merriam-Webster.com
 Sourcers and recruiters need technology that 
can enable their productivity 
 Intelligent search and match apps are not a 
replacement for creative, curious, investigative 
people 
 Do not seek to automate that which you do not 
understand and cannot accomplish manually!
“Computers move information, 
people do the work” 
- Jeffrey Liker
Sourceconaifullv6forslideshare 120108133422-phpapp02

Sourceconaifullv6forslideshare 120108133422-phpapp02

  • 1.
    Talent Sourcing andMatching Artificial Intelligence & Black Box Semantic Search vs. Human Cognition & Sourcing Glen Cathey www.linkedin.com/in/glencathey www.booleanblackbelt.com
  • 2.
    What’s the bigdeal anyway? Some people believe resume, LinkedIn and Internet sourcing is so easy that sourcing is either dying or dead or can be performed for $6/hour
  • 3.
    Resume and LinkedInsourcing appears simple and easy on the surface, however – it is deceptively difficult and complex
  • 4.
     Anyone canfind candidates because all searches "work" as long as they are syntactically correct  That doesn’t mean the searches are finding all of the best candidates!  People make assumptions when creating searches  Every time an assumption is made, there is room for error and you unknowingly miss and/or eliminate results!
  • 5.
     No singlesearch can return all potentially qualified people  Every search both includes some qualified people and excludes some qualified people  Some of the best people have resumes or social profiles that may not appear to be obvious or strong matches to your needs
  • 6.
     People cannoteffectively be reduced to and represented by a text-based document  Job seekers are NOT professional resume or LinkedIn profile writers  Most people still believe shorter and concise resumes and social profiles are still better  This means they are removing data/info from their resumes which can no longer be searched for!
  • 7.
     No onementions every skill or responsibility they’ve had, nor describes every environment they’ve ever worked in  There are many ways of expressing the same skills and experience  Employers often don’t use the same job titles for the same job functions
  • 8.
     People don’tcreate their resumes and LinkedIn profiles thinking about how you will search for them  Sometimes people don’t even use correct terminology  Anyone easy for you to find is easy for other recruiters to find = no competitive advantage!
  • 9.
    In addition tothe people you do find, there are Dark Matter results of people that exist to be retrieved, but can't be found through standard, direct or obvious methods I estimate Dark Matter to be at least 50% of each source searched
  • 10.
  • 11.
    Finding all ofthe best people IS NOT!
  • 12.
    “When every businesshas free and ubiquitous data, the ability to understand it and extract value from it becomes the complimentary scarce factor. It leads to intelligence, and the intelligent business is the successful business, regardless of its size. Data is the sword of the 21st century, those who wield it well, the Samurai.” -Jonathan Rosenberg, SVP, Product Management @ Google
  • 13.
     Stop wastingtime trying to create difficult and complex Boolean search strings  Let "intelligent search and match applications" do the work for you  A single query will give you the results you need - no more re-querying, no more waste of time!
  • 14.
     Understand titles,skills, and concepts  Automatically analyze and define relationships between words and concepts  Intuit and infer experience by context
  • 15.
     Perform patternrecognition  Employ semantic search  Perform fuzzy matching
  • 16.
    How do theyreally work?
  • 17.
     Intuit experienceby context = resume parsing  Parsing breaks down and extracts resume information  Most recent title and employer  Skills and experience  Years of experience – overall, in each position, with specific skills, in management, etc.  Education
  • 18.
     Parsing enablesstructured, fielded search  Search by:  Most recent title  Recent experience  Years of experience  Etc.
  • 19.
     Well developedontologies and taxonomies  Hierarchical
  • 20.
     Synonymous terms  Programmer, Software Engineer, Developer  Tax Manager, Manager of Tax  CSR, Customer Service Representative  Ruby on Rails, RoR, Rails, Ruby  Oracle Financials, Oracle Applications, e-Business Suite, etc.
  • 21.
     Some applicationsuse complex statistical methods in an attempt to "understand" language and the relationships between words  Example: Google Distance
  • 22.
     Keywords withthe same or similar meanings in a natural language sense tend to be "close" in units of Google distance, while words with dissimilar meanings tend to be farther apart
  • 23.
     A measureof semantic interrelatedness derived from the number of hits returned by the Google search engine for a given set of keywords
  • 24.
     Non-interactive andunsupervised machine learning technique seeking to automatically analyze and define relationships between words and concepts  Clustering is a common technique for statistical data analysis
  • 25.
     The designand development of algorithms that allow computers to evolve behaviors based on empirical data  A major focus is to automatically learn to recognize complex patterns and make intelligent decisions and classifications based on data
  • 26.
     Aims toclassify data (patterns) in resumes based either on a priori knowledge or on statistical information extracted from the patterns  A priori: independent of experience  Example of pattern recognition: spam filters
  • 27.
     Finds approximatematches to a pattern in a string  Useful for word and phrase variations and misspellings
  • 29.
     Reduce timeto find relevant matches  Can lessen or eliminate the need for recruiters to have deep and specialized knowledge within an industry or skill set  Reduce and even eliminate time spent on research
  • 30.
     Go beyondliteral, identical lexical matching  Levels the playing field  Can make an inexperienced person look like a sourcing wizard  Good for teams with low search/sourcing capability
  • 31.
     Work wellfor positions where titles effectively identify matches and where there is a low volume and variety of keywords  Good for a high volume of unchanging hiring needs
  • 32.
     Removes thoughtfrom the talent identification and decision making process  Danger of eliminating the need for recruiters to understand what they’re searching for  Information technology, healthcare, and other sectors/verticals can create pose serious challenges to matching apps
  • 33.
     Apps findsome people, bury or eliminate others  Is finding some people good enough for your organization?  Shouldn’t your goal be to find ALL of the BEST people?
  • 34.
     Matching appslevel the playing field  People from different companies using the same solution will both find and miss the same people  Competitors using the same search and match solution will have no competitive advantage over each other!
  • 35.
     Belief thatone search finds all of the best candidates is intrinsically flawed and simply not based in reality  Top talent isn't represented by what a search engine "thinks" has the best resume or profile  AI and semantic search apps favor keyword rich resumes and profiles
  • 36.
     Keyword poorresumes and profiles may in fact represent better talent than keyword rich resumes and profiles  It’s not just a matter of keyword frequency or even keyword presence!  AI powered search & match applications can only return results that explicitly mention required keywords and their variants
  • 37.
     Many peoplehave skills and experience that are simply not mentioned anywhere in their resumes!  These people are the Dark Matter of databases, ATS’s, and social networks, and they exist but cannot be found via direct search/match methods – AI or otherwise!
  • 38.
     Pre-built taxonomiesare static, limited in their completeness and must be continually updated in order to stay relevant and effective  Taxonomies are only as good as who created them  Applications can only match on what’s present and cannot “think outside of the box”
  • 39.
     Semantic clusteringand NLP applications can retrieve related search terms, but that does not mean they are relevant for your need!
  • 40.
     Match primarilyon titles and skill terms  True match is at the level of role, responsibilities, environment, etc.  Some applications rank results favoring recent employment duration  Is someone who has been in their current company for 5 years really “better” than someone who has been with their current company for 2 years?
  • 41.
     Apps don’t"know" what you’re looking for or what's the best match for your company  Apps are not and cannot be "aware" of people that were excluded from their search results  Applications are not truly intelligent – they do not actually "know" or "understand" the meaning of titles and terms
  • 43.
     The abilityto learn or understand or to deal with new or trying situations  The ability to apply knowledge to manipulate one’s environment or to think abstractly  REASON; the power of comprehending and inferring Source: Merriam-Webster.com
  • 44.
     The capabilityof a machine to imitate intelligent human behavior  Artificial = humanly contrived Source: Merriam-Webster.com
  • 45.
     Dr. MichioKaku  Theoretical physicist and futurist specializing in string field theory  Harvard Grad (summa cum laude)  Berkeley Ph.D  Currently working on completing Einstein's dream of a unified field theory  What are his thoughts on AI?
  • 46.
     “…pattern recognitionand common sense are the two most difficult, unsolved problems in artificial intelligence theory. Pattern recognition means the ability to see, hear, and to understand what you are seeing and understand what you are hearing. Common sense means your ability to make sense out of the world, which even children can perform.” - Dr. Michio Kaku
  • 47.
     Dr. MichioKaku believes the job market of the future will be “dominated by jobs involving common sense (e.g. leadership, judgment, entertainment, art, analysis, creativity) and pattern recognition (e.g. vision and non-repetitive jobs). Jobs like brokers, tellers, agents, low level accountants and jobs involving inventory and repetition will be eliminated.”
  • 48.
     That’s goodnews for sourcers and recruiters who perform sourcing!  Sourcing requires judgment, creativity, analysis, common sense and pattern recognition (instantly making sense of human capital data)  Sourcers of the future will be human capital data analysts who are experts in HCDIR & A – Human Capital Data Information Retrieval and Analysis
  • 49.
     Matching appsdo not have the dynamic ability to learn, understand and instantly relate new concepts and through direct experience and observation  They depend on taxonomies, statistical models, or semantic clustering to “understand” relationships and concepts
  • 50.
     The humanmind naturally organizes its knowledge of the world, instantly relating new terms and concepts and judging their relevance
  • 51.
     Example: Asourcer who is completely unfamiliar with “infection control” can instantly recognize non-highlighted but related and relevant terms and incorporate them into new and improved searches Carolinas HealthCare System, Charlotte, NC Infection Preventionist 1997-present Responsible for all aspects of infection prevention and control for an 800+bed hospital. Uses science-based research to perform infection prevention. Conducts all aspects of surveillance, data analysis, and presents data to interdisciplinary teams, including the Infection Control Committee.
  • 52.
     Human sourcerscan learn from research and search results, dynamically and adaptively identifying related and relevant search terms and incorporate them into successive searches to continuously refine and improve searches for more relevant results
  • 53.
     For example,if a recruiter was sourcing for a position that required a skill that they were unfamiliar with (e.g.,“Cockburn Use Case Methodology” ) they could quickly perform research to learn more about it  In the next slide, you will see a screen capture of such research
  • 54.
     From thisquick research , the recruiter would be able to determine that most people would not explicitly mention “Cockburn Use Case Methodology,” let alone “Cockburn” (which the research revealed is pronounced “Co-burn”) – thus they would not include the term in their searches
  • 55.
     Instead, itwould be a better idea to search for candidates that mention experience with Agile methodology and simply call and ask them if they have experience with using Cockburn’s use case methodology (which many likely would)
  • 56.
     Applications usingNatural Language Processing do not truly understand human language  They use complex statistical methods to resolve the many difficulties associated with making sense of human language  NLP experts admit that to computers, even simple sentences can be highly ambiguous when processed with realistic grammars, yielding thousands or millions of possible analyses
  • 57.
     Humans effortlesslyand automatically process and understand language, regardless of sentence length or complexity, ambiguity, incorrect grammar, etc.  We can udnretsnad any msseed up stnecene as lnog as the lsat and frsit lteetrs of wdros are in the crrcoet plaecs
  • 58.
     Human sourcersand recruiters can deduce potential experience, even in the absence of information (not explicitly mentioned in the resume/profile)  Applications can only work with what’s actually mentioned in a resume – if it's not explicitly mentioned, it can't match on it
  • 59.
     Applications arenot aware that many of the best people have average resumes  Applications are not aware of the people their algorithms bury in results or eliminate entirely  Human sourcers can become aware of and specifically target this Dark Matter
  • 60.
    How can youtarget resumes and LinkedIn profiles that exist, but your searches can’t and don’t retrieve them?
  • 61.
     Well developedtaxonomies, semantically generated query clouds and matching algorithms can help greatly with automatically searching for and matching on synonymous terms, related words, word variants, misspellings, etc.
  • 62.
    Think + PerformResearch  For keyword, phrase or title you are thinking of using in your search, realize: 1. Not everyone will explicitly mention what you think they would or should mention in their resume/profile 2. There are many different and often unexpected ways of expressing the same skills and experience
  • 63.
    Global Experience Whatsearch terms might you use if you are looking for people with global experience? How many can you think of off the top of your head?
  • 64.
    In a fewminutes of exploratory research, a sourcer can come up with a volume of related and relevant terms  Global, international, foreign, multinational, worldwide  Europe, European, EU, EMEA, Asia, Asia-Pac, Pacific Rim, South America, Latin America, Americas, CALA (Caribbean and Latin America), Middle East  Canada, Japan, China, Russia, India, UK, United Kingdom, etc.  Countries, Offshore, Overseas
  • 65.
    How can youtarget results of people that your searches retrieve but the results are buried (ranked poorly or "too many" results to be reviewed) and you don’t find them?
  • 66.
     Search andmatching software powered by artificial intelligence / black box semantic search doesn't have a solution to this challenge  One of the major claims AI/semantic search applications make is that their solutions can find the "right people" in one search
  • 67.
     However -a single search strategy is intrinsically flawed and limited - no single search can find all qualified candidates, and each search both includes qualified people as well as excludes qualified people  I am not aware of any search & match software that allows for successive searching via mutually exclusive filtering
  • 68.
    Run Multiple Searches  Start with maximum qualifications  Use the NOT operator to systematically filter through mutually exclusive result sets  End with minimum qualifications
  • 69.
     Required: A,B,C  Explicitly desired: D,E  Implicitly desired: F
  • 70.
    1. A andB and C and D and E and F 2. A and B and C and D and E and NOT F 3. A and B and C and D and NOT E and F 4. A and B and C and NOT D and E and F 5. A and B and C and NOT D and NOT E and F 6. A and B and C and D and NOT E and NOT F 7. A and B and C and NOT D and E and NOT F 8. A and B and C and NOT D and NOT E and NOT F
  • 71.
  • 72.
    Probability-Based and Exhaustive! This approach allows for: 1. The specific targeting of people who theoretically have the highest probability of being a match based on information present 2. The specific targeting of people who may be the best match, but may have keyword/information poor resumes or profiles, who do not explicitly mention what you think the "right" person would or should mention 3. The ability to systematically filter through all available results via manageable and mutually exclusive result sets – never seeing the same person twice!
  • 74.
     A mixof “man and machine,” integrating human knowledge and expertise into computer systems  Essentially - the best of both worlds:  Autopilot: An artificially intelligent semantic matching engine  Manual Override: Ability to take complete control over searches and search results
  • 75.
     An artificialintelligence semantic matching engine coupled with taxonomies built by human SMEs that are continually modified and improved specifically for the organization  No COTS solution is customized for any specific employer, industry or discipline, nor 100% complete
  • 76.
     Resume andLinkedIn profile parsing  Structured, contextual search  Most recent title and experience, overall years of experience, education, etc.  White Box relevance weighting  Configurable by users – no black box!  Searchable tagging for level 5 semantic search
  • 77.
     Standard andextended Boolean in full text and field-based search  AND, OR, NOT, configurable proximity, weighting  Configurable proximity enables level 3 semantic search  Variable term weighting allows users to control which search terms are more important and thus control over true relevance
  • 78.
     Lucene isa free and open source text search engine that support configurable proximity and term weighting, and can be integrated into some existing ATS's/databases  Some Applicant Tracking Systems already have databases powered by text search engines that allow for extended Boolean
  • 79.
    “Society has reachedthe point where one can push a button and immediately be deluged with…information. This is all very convenient, of course, but if one is not careful there is a danger of losing the ability to think.” - EijiToyoda
  • 80.
     Data andinformation requires analysis to support decision making  Just as very expensive Business Intelligence and Financial Analytics software hasn't replaced the need for people to make sense of the data, there is no software solution for HR and recruiting that replaces the need for people to analyze and interpret human capital data to make appropriate decisions
  • 81.
     Matching appsmove/retrieve information, but only PEOPLE can analyze and interpret for relevance and make intelligent decisions  Relevant: the ability (as of an information retrieval system) to retrieve material that satisfies the needs of the user [1]  Only the user (sourcer/recruiter) can judge relevance! [1] Source: Merriam-Webster.com
  • 82.
     Sourcers andrecruiters need technology that can enable their productivity  Intelligent search and match apps are not a replacement for creative, curious, investigative people  Do not seek to automate that which you do not understand and cannot accomplish manually!
  • 83.
    “Computers move information, people do the work” - Jeffrey Liker

Editor's Notes

  • #10 Lost and Found by metrognome0 via Flickr/creative commons
  • #11 Haystack image
  • #12 Haystack image
  • #15 Concept = meaningful combination of words
  • #18 Decrease content and speak to?
  • #20 Hierarchical, parent-child, one-way: ontologies apply a larger variety of relation types/categorization Practice and science of classification Healthcare - hospital
  • #23 Source: Wikipedia Luwig Wittgenstein’s theories about how words are defined by context
  • #24 Source: Wikipedia Luwig Wittgenstein’s theories about how words are defined by context
  • #25 Query clouds
  • #26 Scientific discipline
  • #27 A priori: independent of experience – book learning vs. OJT A posteriori: dependent on experience or empirical evidence
  • #34 Some people = good enough for your organization? Find the best, and ALL of the best Means the same candidates are also missed
  • #40 Statistical NLP - Automatically analyze and define relationships between words and concepts Relevant: the ability (as of an information retrieval system) to retrieve material that satisfies the needs of the user
  • #43 http://www.flickr.com/photos/stickergiant/4793776078/sizes/o/in/photostream/
  • #44 Infer = to derive as a conclusion from facts or premises
  • #45 More accurately, created intelligence
  • #46 B.S., summa cum laude from Harvard University Ph.D, University of California, Berkeley
  • #47 Those two problems are at the present time largely unsolved. Now, I think, however, that within a few decades, we should be able to create robots as smart as mice, maybe dogs and cats.
  • #48 Sourcing requires creativity, interpretive analysis, judgment, and common sense – a natural understanding based on experience
  • #49 Sourcing requires creativity, interpretive analysis, judgment, and common sense – a natural understanding based on experience
  • #50 Dynamic – continuous and productive activity or change Static – showing little/no change
  • #51 Dynamic Inference = Infer = to derive as a conclusion from facts or premises
  • #52 Dynamic Inference = Infer = to derive as a conclusion from facts or premises
  • #53 Dynamic Inference = Infer = to derive as a conclusion from facts or premises
  • #54 Dynamic Inference = Infer = to derive as a conclusion from facts or premises
  • #55 Dynamic Inference = Infer = to derive as a conclusion from facts or premises
  • #56 Dynamic Inference = Infer = to derive as a conclusion from facts or premises
  • #58 Dynamic – continuous and productive activity or change
  • #60 Aware: having or showing realization, perception, or knowledge What if? How else? Curious…
  • #75 The system gives you what it thinks you wanted, and you are able to tell the system what you wanted A Knowledge Engineering system that integrates human knowledge into computer systems in order to solve complex problems normally requiring a high level of human expertise
  • #76 An expert system is computer software that attempts to mimic the reasoning of a human specialist
  • #80 Kaizen and the Toyota Way
  • #81 Financial data/BI - Tons of software – financial EPR, business intelligence apps – they STILL require people to analyze and interpret Applications can’t truly analyze/interpret
  • #82 Financial data/BI - Tons of software – financial EPR, business intelligence apps – they STILL require people to analyze and interpret Applications can’t truly analyze/interpret