Back in March 2010, I had the distinct honor of delivering the keynote presentation at SourceCon on the topic of resume search and match solutions claiming to use artificial intelligence in comparison with people using their natural intelligence for talent discovery and identification.

Now that nearly 2 years has passed, and given that in that time I’ve had even more hands-on experience with a number of the top AI/semantic search applications available (I won’t be naming names, sorry), I decided it was time to revisit the topic which I am very passionate about.

If you’ve ever been curious about semantic search applications that “do the work for you” when it comes to finding potential candidates, you’re in the right place, because I’ve updated the slide deck and published it to Slideshare. Here’s what you’ll find in the 86 slide presentation:

• A deep dive into the deceptively simple challenge of sourcing talent via human capital data (resumes, social network profiles, etc.)
• How resume and LinkedIn profile sourcing and matching solutions claiming to use artificial intelligence, semantic search, and NLP actually work and achieve their claims
• The pros, cons, and limitations of automated/black box matching solutions
• An insightful (and funny!) video of Dr. Michio Kaku and his thoughts on the limitations of artificial intelligence
• Examples of what sourcers and recruiters can do that even the most advanced automated search and match algorithms can’t do
• The concept of Human Capital Data Information Retrieval and Analysis (HCDIR & A)
• Boolean and extended Boolean
• Semantic search
• Dynamic inference
• Dark Matter resumes and social network profiles
• What I believe to be the ideal resume search and matching solution

I’ve been blogging nearly 3 years now, and I realized I’ve never come out and actually defined the term ”Boolean Black Belt.”

The concept seems pretty self explanatory, but there has been at least 1 person who’s taken the opportunity to point out (and gain some traffic in the process – but it’s all good!) that it could be perceived as a bit of an oxymoron to be an “expert” in something as simple as 3 Boolean operators.

Interestingly, however, I’ve found that most sourcers and recruiters don’t even fully exploit the various powers of the OR and NOT operators – not even close.

So what is a “Boolean Black Belt” anyway?

Black Belt

I use the term ”Black Belt” in reference to the widely known way of describing an expert in martial arts, where the black belt is commonly the highest belt color used and denotes a high degree of competence.

That’s the easy part; the “Boolean” part isn’t so simple to define.

Boolean

I’d like to take the opportunity to clear up some misconceptions about, and disambiguate my use of “Boolean” in “Boolean Black Belt,” and pretty much any article in which I refer to Boolean.

When I refer to “Boolean,” I am not refering only to the basic Boolean operators of AND, OR, and NOT. I’m actually referring to the entire process of:

1. Analyzing, understanding, and interpreting job opening/position requirements
2. Taking that understanding and intelligently selecting titles, skills, technologies, companies, responsibilities, terms, etc. to include (or purposefully exclude!) in a query employing appropriate Boolean operators and query modifiers
3. Reviewing the results of the initial search to assess relevance as well as scanning the results for additional and alternate relevant search terms, phrases, and companies
4. Based upon the observed relevance of and intel gained from the search results, modifying the search string appropriately and running it again
5. Repeat steps 3 and 4 until an acceptably large volume of highly relevant results is achieved

Instead of trying to put all of that into a domain name and a concise catch phrase, hopefully you can appreciate why I chose to summarize that entire process as ”Boolean.”

Beyond Boolean Logic

Admittedly, the basic Boolean operators are easy to learn – after all, there’s only 3 of them!

However, anyone who’s adept at leveraging databases and information systems for talent identification knows that the “magic” does not lie in the operators themselves, but in all of the steps detailed above.

The “real” work of creating effective Boolean search strings lies in the interpretive analysis of the need, determining what terms to include and exclude from searches and in what specific combination, in the analysis of the relevance of the initial search results, and the adaptive process of learning from the results to further refine the Booleans to find a large quantity of highly relevant results – people who are highly likely to be (or know!) the right match for your hiring needs.

What I just described is actually the process of Information Retrieval (IR), but no matter how much I write on the subject, people still cling to “Boolean.”

Sourcing isn’t so Simple

While learning about the concepts of basic Boolean logic is easy, there is nothing inherently easy about creating Boolean search strings for talent identification.

To say that searching databases and information systems to identify talent is “easy” because it’s defined only by 3 simple Boolean operators is to admit that you have little to no understanding or appreciation of online, database, or social network sourcing.

That would be like saying that a challenging math-based brain teaser is simple because everyone understands addition, subtraction, division, and multiplication.

For example, this classic puzzle should be easy for anyone who understands basic math, right?

“My grandson is about as many days as my son is weeks, and my grandson is as many months as I am in years. My grandson, my son and I together are 100 years. Can you tell me my age in years?”

After all, it only requires 3 basic and simple mathematical operations: addition, multiplication, and division. If that one is too “easy” for you, give this brain teaser a try – it too only requires basic math to solve.

It should be obvious that the real challenge of math-based problems comes from being able to understand the puzzle in the first place, and then determining precisely what types of equations and operations are required to solve the problem.

The analysis and understanding is primary, the mathematical operators secondary, as they are useless without the proper understanding of the required and specific application of them.

It’s the same thing with Boolean search strings.

Extended Boolean

Beyond the 3 “standard” Boolean operators, there lies extended Boolean, which typically includes proximity operators and term weighting/boosting.

While not every search engine supports extended Boolean, those that do afford users the ability to dramatically increase the relevance of search results, effectively enabling user-defined semantic search.

Semantic Search

Semantic search can be defined as search techniques that leverage the actual meaning in words and phrases and can return results that more closely match the “meaning”  or intent of the search rather than simply returning results that match the words of the search.

The whole goal of searching databases, the Internet, social media, or other information systems is ostensibly to find people who have a high likelihood of being (or knowing!) a potential match for a hiring need that you have now, or will have in the future.

The more skill and ability you have in being able to craft and execute Boolean and extended Boolean search strings that find more of the right people more quickly, the more effective you can be as a Sourcer or Recruiter.

By “effective” I mean filling more positions with high quality talent while reducing time-to-fill.

More. Faster. Better.

Whenever I refer to “Boolean” in articles or even in the name of this blog, I’m actually referring to extended Boolean and user-defined semantic search as well as the basic Boolean operators.

Query Modifiers

Boolean search strings are often comprised of more than just search terms and Boolean operators.

There are also query modifiers, and depending on the search engine, they can include: *, ” “, inurl:, ~, ( ), w/, and many more.

Anyone hoping or claiming to have a high degree of competence with sourcing not only has to have a solid command of the basic Boolean operators, but also how to leverage the available and appropriate query modifiers.

Final Thoughts

I use the term “Boolean Black Belt” to describe someone with a high degree of competence in the entire process of interpreting and understanding a specific talent need, determining what terms to include and/or exclude from searches and in what specific combination, crafting search strings making effective and appropriate use of Boolean operators, query modifiers, search terms, and semantic search techniques, the analysis of the relevance of the initial search results, and the adaptive process of learning from the results to further refine the Booleans to find a large quantity of highly relevant results – people who are highly likely to be (or know!) the right match for their hiring need.

I believe that when most people in sourcing and recruiting roles refer to “Boolean,” they are not simply referring to AND, OR, and NOT.

To say that mastering the use of Boolean search strings for talent identification is limited to the understanding of the functions of 3 Boolean operators would be ridiculous and an obvious sign of ignorance.

Most people would agree that Barack Obama is an excellent orator, yet he does not use words most people do not understand. For the most part, he uses common words that everyone is familiar with. But his ability as an orator cannot be defined by or limited to the common words he uses - it lies in how he organizes his thoughts and how he arranges and delivers his sentences to convey his indended meaning.

Most sculptors, golfers, jiu jitsu practitioners, and orators use the same tools, clubs, moves, and words. However, mastery does not come from the specific tools, clubs, movements, or words - it’s in the appropriate and effective APPLICATION of them, typically in response to a challenge or to achieve a specific goal.

Knowing what golf clubs are and how to swing them does not make you a world-class golfer. Having a good vocabulary does not make you an excellent public speaker. Knowing how to punch and kick will not ensure you can win any martial arts/MMA competitions. Owning a hammer and chisel does not make you a world-renowned sculptor.

Similarly, having a command of 3 Boolean operators does not ensure that you can understand the positions you are sourcing or recruiting for and effectively leverage electronic sources of human capital data (databases, ATS/CRM’s, social media, the Internet, job boards, etc.) to find more of the best candidates available for your hiring needs more quickly.

Nor does it define a Boolean Black Belt, if such a thing can or should exist.

Beyond Basic Boolean

Most sourcing, recruiting, and staffing professionals are familiar with the basic Boolean operators of AND, OR, and NOT. However, I have found that few are familiar with what some refer to as “extended” Boolean functionality, such as proximity search and term weighting.

Proximity and term weighting, where supported, are not actually logical (Boolean) operators – they are more accurately referred to as text or content operators.

Whatever you call them – extended Boolean or text operators – they offer sourcers and recruiters significantly more control, power and precision when executing searches, and in the hands of an expert, they can enable semantic search.

Relevance is Everything!

When it comes to search – relevance rules.

Ultimately, any sourcing or recruiting professional knows that what’s most critical in running Boolean searches on LinkedIn, the Internet, a job board, or in an internal resume database is getting relevant results.

However, few people talk about exactly what determines relevance – and I think I know why.

According to Wikipedia, “relevance” denotes how well a retrieved set of documents (or a single document) meets the information need of the user.

The problem is that no search engine, social networking site, or database can “know” what is relevant to you – only you can determine how relevant results are because only you know what you were looking for in the first place!

For sourcing and recruiting, relevant results are typically defined as resumes or profiles of (or information about) potential candidates whose experience and capabilities closely match the hiring profile or job opening that the sourcer or recruiter is trying to find candidates for.

I’d argue that the value of any source of information (LinkedIn, a resume database, the Internet, etc.) lies less in the information contained within, and more in the ability of a user to extract out precisely and completely what the user needs – finding and retrieving any and all appropriately qualified candidates.

Information has no value to you if you are unable to find it and take action on it.

So how can extended Boolean help sourcers and recruiters find more relevant results?

Let’s take a look at proximity first.

Proximity Search

Proximity search functionality enables a user to search for specific terms that are mentioned within a certain distance of other specific terms.

Being able to control how close search terms are to each other can be especially helpful when leveraging the structure of certain websites and pages – I’ll demonstrate this later in the post using LinkedIn and Twitter as examples.

In my opinion, the more powerful application of proximity search lies in the ability to perform natural language or semantic search.

Semantic search uses the science of meaning in language to produce highly relevant search results rather than have a user sort through a list of loosely related keyword results. Words that are close together are often in the same sentence, and when you can search for meaning at the sentence level, you can target people based on what they actually do/what their responsibilities have been.

Being able to target sentences in which people detail their specific responsibilities and level of responsibility is absurdly more powerful than basic keyword search (Level 1 Talent Mining), which is prone to low levels of relevance and false positives.

There are 3 main types of proximity searching: fixed proximity, variable proximity, and adjacency. For the purposes of this post – I will focus only on fixed and variable proximity.

Fixed Proximity Search

Fixed proximity is most commonly represented by the NEAR operator. The search engines that do recognize and support the NEAR operator typically define NEAR proximity as within 1 to 10 words (specific search engines can differ – check their documentation). Monster’s resume database supports the NEAR operator (which doesn’t have to be capitalized, btw) at a fixed distance of up to 10 words.

How could you leverage fixed proximity to find more relevant search results?

If you were looking for a Windows and Exchange administrator, any basic keyword and title search can pull tons of results of resumes that mention all of the search terms, as well as a high percentage of false positive results. False positive results in this example would be of resumes that mention all of the search terms and titles, but the people have never been primarily responsible for administering windows and exchange servers. A 1 year helpdesk professional can show up in these results because all they have to do is mention the keywords somewhere in their resume.

Leveraging fixed proximity, you could craft this (purposefully basic) search using the NEAR operator: Windows and Exchange NEAR admin* and server*.

That search will ONLY return results of resumes/profiles that mention Exchange within 1 to 10 words of any word starting with the root of admin (administrator, administration, administer, administered, etc.).

Being able to control the fact that Exchange MUST be mentioned within close proximity to admin* will dramatically affect and improve the relevance of the search results, typically returning results of candidates who either have a title using both terms and/or candidates that talk about being responsible for Exchange administration.

• Managed & administered more than 300 Exchange Servers
• Provisioned & administer multiple Exchange 5.5/2003 servers
• Not only are there administration duties for Exchange and Blackberry…
• Exchange/RightFax administrator
• Installing, Configuring, and Administering Microsoft Exchange 2000 Server
• Administer a Microsoft Exchange 2003/2007 environment
• 8+ years of expertise as a System Administrator in Windows 2003 family, Windows 2000 family, MS Exchange 5.5, MS Exchange 2000, and Exchange 2003
• I am proficient with the following skills; planning, installation and administration of Windows Active Directory, Windows Servers, Exchange Server
• Windows Server Support, Active Directory,Exchange Server 2000, 2003 administration and Blackberry Server administration
• Administer Exchange 2003 Server for corporate email

As you can see, being able to control the proximity of specific search terms essentially increases the likelihood of returning results of candidates who have had administrative responsibility for Exchange servers, effectively increasing the relevance of the results, because that’s what we were actually trying to find and identify!

Configurable Proximity

A search engine that supports configurable proximity affords users the ability to precisely control the distance between specific search terms.

This can produce even more relevant results than the NEAR operator, because the NEAR operator’s maximum range of 10 can allow for some non-relevant results to be returned. The farther words are mentioned apart from each other, the less likely it is that they are semantically related. In fact, at a distance over 10 words, each word could easily be mentioned in separate bullet points or in separate sentences on a resume and be completely unrelated.

However, with configurable proximity, a sourcer or recruiter can choose the maximum distance between search terms.

Instead of being limited to a distance of 10 or fewer words, a search engine that allows for configurable proximity allows you to create searches that force terms to be quite close together – as close as you like.

For example, you could choose to search for only people who mention Exchange within 5 words of any word starting with the root of admin (administrator, administration, administer, administered, etc.), regardless of order. A maximum distance of 5 words will dramatically increase the relevance of the search results because mentioning those 2 search terms at such a close range makes it more likely that they are mentioned in the same bullet point or sentence and thus more likely to be semantically related.

Essentially, this search will only return results of people who specifically mention something about being responsible for administering Exchange at least once in their resume. By employing this kind of search, a sourcer is actually performing a semantic search, targeting sentence-level meaning, as they are looking specifically for people who talk about having a particular responsibility – not just looking for documents that happen to contain the search terms.

Leveraging Website and Page Structure with Proximity Search

Once you have noticed a consistent pattern to the structure of certain websites and pages, you can use Internet search engines that support proximity search to target the distance between search terms to yield highly relevant search results.

Although Google supposedly supports proximity search with their undocumented AROUND(x) search operator, I have found its reliability to be suspect. Perhaps that’s why it’s not officially documented?

The good news is that Bing’s configurable proximity search functionality of NEAR:x seems to work quite well and consistently.

To leverage the structure of certain websites such as LinkedIn, here is a quick example of how you can target current titles and companies when using Bing.

site:linkedin.com current near:3 “engineer at Google” “san francisco bay area”

In this query, all of the results must have the phrase “engineer at Google” within 3 words of the word “Current,” which is on every LinkedIn profile.

If you click on any of the cached results, you can see how Bing happily returned results of people who have the phrase “engineer at Google” in their current title field:

With Bing’s NEAR:x functionality, it is remarkably simple to X-Ray Twitter and target people in specific locations who mention specific titles and/or skill terms in their bios.
For example, let’s say you wanted to find Twitter profiles of user experience professionals who live in the New York area. You could run a search like this on Bing to force the search engine to return only results that mention UX within 15 words of “Bio” and “New York” within 3 words of “Location:”

site:twitter.com bio near:15 UX location near:3 new york

You can see how Bing’s proximity search helps you target terms in Twitter bios and location text:

Viewing a cached result displays Bing’s NEAR:x flawless execution:

How’s that for a relevant result?

Basically as good as it gets – I wanted someone who lives in the NY area who is a User Experience professional, and that’s exactly what I got! That is relevance!

Of course, when searching Twitter, it is especially important to realize that people can be very creative in how they may describe themselves (titles, skills, etc.), their experience, and their location – they can enter whatever they want.

As such, you could not find the above Twitter bio by searching only for “Drupal.”

Performing Semantic Search with Configurable Proximity

You can perform basic semantic search by targeting sentence-level meaning using Bing’s support of configurable proximity.

For example, let’s say you were searching for resumes on the Internet and wanted to find people who have had a specific responsibility, such as configuring juniper routers.

You could run a basic search like this: (inurl:resume OR intitle:resume) configuring near:5 juniper juniper near:5 routers

And see results like this:

Of course, there are many different ways to run that search – I only wanted to demonstrate the power of being able to control how close search terms are to each other, especially when targeting responsibilities, typically stated in verb/noun combinations. This allows you to perform semantic search at the sentence level.

Now that we’ve played around a bit with proximity search, let’s move onto the other half of extended Boolean – variable term weighting.

Variable Term Weighting

Talented sourcers and recruiters know that not all terms are equally important in a query.

In most queries and searches, certain search terms are more important than others. When running standard Boolean queries, all search terms are considered/weighted equally – and this is the stone that the makers of so-called semantic search applications often throw at Boolean search.

Unfortunately, many search engines and database search interfaces simply assign relevance to results by the number of search term “hits” in each document. In most cases, the simple frequency of search terms does not correlate to relevant results. This is where the derisive description “buzzword bingo” comes from, most often used to denote that there is little skill involved in running Boolean searches counting matched keywords.

Using an Information technology hiring profile as an example – if a sourcer was looking for candidates who have significant experience administering Windows servers and Exchange email servers they might create a simple Boolean query such as this: Windows AND Exchange AND server* and admin*.

That search is highly likely to return and rank candidates who are Windows systems administrators who mention Windows many times in their resume/profile and happen to mention Exchange once or twice as highly relevant because of the number of “hits” for Windows – which is by nature a very common term in resumes.

This would leave the sourcer with having to sort through a large volume of false positive results (that contain the keywords, but are not of people who have been primarily responsible for administering Windows and Exchange servers) to find the candidates who actually have been primarily responsible for administering Exchange servers as well as Windows servers.

Search engines that offer users the ability to assign different weights to each search term enable sourcers and recruiters to move beyond simple buzzword matching and take control of the relevance of the results. Essentially, with variable term weighting you can assign a number value to words to increase their weight when ranking retrieved documents – which does not change the total number of results, but the ORDER of the results.

Using the same example as above, a sourcer using a search engine that supports variable term weighting could create a Boolean search string to more heavily weight the term “Exchange.” That Boolean query would pull the same number of results as the first search that had no term weighting – however, it would sort and rank the results heavily favoring resumes/profiles that mention Exchange more often in relation to the other search terms, increasing the likelihood that the sourcer can quickly identify candidates who have had experience being responsible for administering and supporting Exchange servers.

By employing variable term weighting, you can positively affect the relevance of the search results.

Final Thoughts

Hopefully I’ve shed some light on how being able to control the proximity of two search terms can yield results that are FAR more relevant than results that simply mention the two terms anywhere in a document or form – this is the critical difference between the semantic similarity between a search and its results vs. the lexical similarity between a search and its results.

There are countless ways you can apply extended Boolean functionality such as variable term weighting and proximity searching to nearly any industry/hiring profile to create searches that return highly relevant results - results that are more relevant than those that can be achieved with standard Boolean logic.

Using a search engine that supports both variable proximity and variable term weighting can empower sourcers and recruiters to quickly find large volumes of highly relevant results, increasing productivity and achieving Just-In-Time sourcing and recruiting.

I wish the makers of search engines would seek less to “dummy-down” search interfaces and functionality and incorporate more powerful search capability that allows users to take significant control over the relevance of their search results.

There are a few search engines and ATS/CRM systems that support both configurable proximity search and variable term weighting.

Does yours?

I’ve been a Google search fan for many years – since 1998, and I’ve used it exclusively for all of my search needs, both personal and professional.

Until recently.

That’s because I’ve discovered that Bing has a number of advantages over Google when it comes to sourcing candidates, including:

• Cleaner, shorter, simpler and effective LinkedIn X-Ray searching
• Effective Twitter X-Ray searching
• Never doubting your humanity and refusing to run your more advanced queries
• Configurable proximity (although I just learned Google has a similar capability)
• Converting searches into RSS feeds

X-Ray Searching LinkedIn

I’ve been hacking away at public (and “private“) LinkedIn profiles for quite some time, and exclusively using Google to do so until a few months ago.

Once I started playing around with Bing to search LinkedIn, I quickly found out that Bing isn’t prone to refusing to run your searches like Google does when they politely inform you that they’re sorry, “but your computer or network may be sending automated queries. To protect our users, we can’t process your request right now.” In fact, Bing has never thought my LinkedIn X-Ray searches were malicious automated queries. Thanks Bing!

Additionally, Bing lets you specifically target LinkedIn profiles by simply adding “powered” to your search strings, instead of all of the things you need to include in a Google X-Ray search of LinkedIn to isolate profiles from other types of undesirable results, such as (inurl:pub | inurl:in) -intitle:directory -inurl:dir -inurl:jobs, and so on.

Bing also makes it very easy to find the full names of most “private” profiles and 3rd degree connections (for those using LinkedIn with a free account). In many cases, all you need to do is copy and paste the “headline” from the LinkedIn profile in question into Bing as a phrase in quotes to get the public profile you’re looking for.

For example:

Taking that headline and searching for it as a phrase in quotation marks on Bing, you will reveal the person’s public profile and full name. Notice that I didn’t have to add site:linkedin.com. If you encounter a profile with a very common/generic headline phrase, such as “Systems Engineer at IBM,” you may not get so lucky.

In cases such as those, all you need to do is use a unique combination of phrases/terms from the profile – any combination that is likely to isolate the specific LinkedIn profile you’re targeting. I find that using the headline, an exact current and/or previous title phrase, and an educational institution works well when the headline alone is very non-specific.

Bing Supports Proximity Search

I’ve written extensively on the topic of the power of proximity search – check out this post on extended Boolean and skip to Level 4 Talent Mining in this post – so I won’t go into deep detail here.

However, I will point out a few cool things you can do with Bing’s NEAR:x command.

When X-Ray searching LinkedIn, you can target current titles and companies when using Bing.

For example: site:linkedin.com powered current near:3 “engineer at Google” “san francisco bay area”

If you click on any of the cached results, you can see how Bing happily returned results of people who have the phrase “engineer at Google” in their current title field:

Bing’s support of configurable proximity search can also be very useful when searching for resumes on the Internet. Let’s say you wanted to find people who have had a specific responsibility, such as configuring juniper routers.

You could run a search like this: (inurl:resume OR intitle:resume) configuring near:5 juniper juniper near:5 routers

And see results like this:

Of course, there are many different ways to run that search – I only wanted to demonstrate the power of being able to control how close search terms are to each other, especially when targeting responsibilities, typically stated in verb/noun combinations. This allows you to perform semantic search at the sentence level.

X-Ray Searching Twitter

With Bing’s NEAR:x functionality, it is remarkably simple to X-Ray Twitter and target people in specific locations who mention specific titles and/or skill terms in their bios.

For example, let’s say you wanted to find Twitter profiles of user experience professionals who live in the New York area. You could run a search like this on Bing:

site:twitter.com bio near:15 UX location near:3 new york

You can see how Bing’s proximity search helps you target terms in Twitter bios and location text:

Viewing a cached result displays Bing’s NEAR:x flawless execution:

Of course, when searching Twitter, it is especially important to realize that people can be very creative in how they may describe themselves (titles, skills, etc.), their experience, and their location – they can enter whatever they want.

Unlike LinkedIn which generates location phrases based on the zip code a user has entered, in Twitter you can can enter anything you want. I’ve seen people who list “Narnia” as their location.

Don’t care about searching Twitter? You should – Twitter has about twice as many users as LinkedIn.

Convert Bing Searches into RSS Feeds

Most sourcers and recruiters are familiar with Google’s email alerts, but I don’t think many are familiar with Bing’s ability to convert searches into RSS feeds.

Using the Bing X-Ray search for engineers at Google from above, here is what you will find in the address bar:

http://www.bing.com/search?q=site%3Alinkedin.com+powered+current+near%3A3+%22engineer+at+Google%22+%22san+francisco+bay+area%22&go=&form=QBRE

To which you can add &format=rss at the end, resulting in this:

http://www.bing.com/search?q=site%3Alinkedin.com+powered+current+near%3A3+%22engineer+at+Google%22+%22san+francisco+bay+area%22&go=&form=QBRE&format=rss

When you add &format=rss and hit “Enter,” you’ll see this:

Notice anything interesting about the results?

Of course, if you use Firefox, you could simply click here to accomplish the same thing:

If you use Internet Explorer, after adding &format=rss and hitting “Enter,” you can add the feed to your Favorites or subscribe to the resulting feed:

Do you think you might have a use for RSS feeds generated from LinkedIn X-Ray searches?

Final Thoughts

I still use Google – old habits die hard, and it’s a good search engine. Even if you are an avid fan and user of Google, Bing cannot be ignored by sourcers and recruiters. It would be folly to not exploit Bing’s many advantages that can be leveraged specifically for talent discovery and identification.

One of my next posts will be a proximity search shootout between Bing and Google, now that I’ve become aware that Google also supports configurable proximity (thanks Kelly!). I will also be exposing some of the limitations of Bing search, including search string length and not properly processing OR statements in certain search scenarios.

Stay tuned!

Did you know that LinkedIn currently has the ability to deliver incredibly powerful search functionality to its users - WELL beyond what we all have access to now?  What am I talking about?

I’m excited to tell you, but quite honestly, I actually can’t believe it’s taken me this long to put 2 and 2 together. Have you ever really watched the video clip below that you can find on  LinkedIn’s Learning Center as well as on YouTube?

If you ignore the information regarding the new features and pay close attention to the video, you can hear Esteban talk about how LinkedIn is always on the lookout for talented Lucene Open Source engineers and watch him search for them. Lucene is an open source text search engine that I’ve written about in multiple posts for its advanced search functionality, including extended Boolean.

LinkedIn uses Lucene as their Text Search Engine

When I first watched the video, I never gave the Lucene stuff a second thought because LinkedIn doesn’t actually offer any of Lucene’s truly advanced search functionality – LinkedIn doesn’t even support root-word/wildcard searching, let alone extended Boolean search. I figured if they were already using Lucene for their text search engine they would offer all of Lucene’s search functionality, which they don’t.

Then I watched the video again the other day (not exactly sure why) and I it made me curious. Had they already implemented Lucene, or were they looking to do so? I did some research to see if I could confirm a link between LinkedIn with Lucene (pun intended).  Although TechCrunch reported that LinkedIn upgraded its people search, they failed to mention the technology behind the upgrade. I was then able to dig up an article that verified that LinkedIn had implemented Lucene as their text search engine.

So What Can LinkedIn Do With Lucene?

I’m glad you asked – be prepared to be amazed! 

Wildcard Searches

Lucene supports single and multiple character wildcard searches within single terms. That means you could search for the term develop* and LinkedIn would return results of people who mention every word that begins with the root of “develop:” develop, developed, developing, developer, develops, etc. That would mean no more having to type out long OR statements where you have to think about all of the different ways a particular term can be written.

Proximity Search

Lucene supports configurable proximity search – or the ability to find words that are a within a specific distance from each other (3 words, 8 words, your choice). For example, if you wanted to find people who mention that they have experience configuring routers, you can use Lucene’s proximity search functionality via the tilde symbol (~) to target phrases where some mention of config* is made within 5 words of router or routers.

“config* rout*”~5

This functionality is HUGE, as it allows sourcers and recruiters to drastically increase the relevance of search results by targeting people based on their responsibilities rather than basic keyword search (aka buzzword bingo). Without forcing some variant of the word ”configure” to be within 5 words of “router” or “routers,” you can just as earily return results of people who do not mention that they have been specifically responsible for configuring routers – you could end up finding people who mention that they’ve configured other things (e.g. servers), and who make 1 mention of the word “router” in their skill summary because they have a router at home (but no paid professional experience). That would be what I call a false positive hit. The result mentioned the search terms, but it did not match the intent of my search – which is to find people who have been responsible for configuring routers.

When I talk about targeting people based on their responsibilities, I mean searching for responsibility verbs (administer, manage, develop, design, configure, filing, reconcile, audit, etc.) mentioned in close proximity (in the same sentence) to skill/technology nouns (oracle, statements, servers, projects, reports, Microsoft Dynamics, SAP, etc.). Being able to control how close words like those are in proximity to each other – down to the sentence level – allows sourcers and recruiters to perform semantic search (aka, natural language search). Essentially, you are able to find people based on what they DO, not just the words they happen to mention in their profile.

If you’re new to the concept of semantic search, I strongly suggest you read these articles (Semantic Search 1, Semantic Search 2, Semantic Search with Proximity, Semantic Search without Proximity) that will throughly explain the concept as well as show you how can currently leverage proximity search to your advantage on Monster and Exalead.

Variable Term Weighting

Here’s the other biggie – Lucene allows you to control the the relevance weighting of your search terms. Lucene calls it “boosting.” In other words – you can tell Lucene that specific terms in your search string are more important/relevant to you than others. That’s right – instead of the search engine taking all of your search terms and “deciding” which results are the most relevant, YOU control the search relevance based on which terms you think are more critical and match the intent of what you’re specifically looking for.

To boost a term with Lucene you can use the caret (^) symbol with a boost factor (a number) at the end of the term you are searching. The higher the boost factor, the more relevant the term will be, so boosting allows you to control the relevance of your results by boosting specific terms.

For example, if you are searching for the following terms: Unix, Windows, Citrix, VMware, storage, and you really needed people who had significant Citrix experience, you can boost that term with the ^symbol:

Unix AND Windows AND Citrix^5 AND VMware AND storage

This will make profiles with more mentions of the term Citrix to appear more relevant and thus be higher in the search results ranking.  This is important, because people who have a lot of experience with Citrix (in terms of specific responsibilities and/or mulitple positions in their career history in which they use Citrix) will likely have multiple mentions of Citrix in their profile. Boosting Citrix will result in bubbling all of the profiles with many mentions of Citrix to the top of the results.

This is especially critical because without the ability to “tell” the search engine with specific terms are actually most relevant to you, the search engine makes its own “decision” as to what’s relevant. And in the case of my example – the search engine may see profiles who mention the word Windows 20 times in their profile as highly relevant, even if they only mention Citrix once – which isn’t likely to actually be someone who matches my need of a strong Citrix professional.

In addition to boosting single terms, you can also boost phrases. By default, the boost factor is 1. Although the boost factor must be positive, it can be less than 1 (e.g. 0.2).

More Lucene Search Functionality

Lucene also supports fuzzy searching (finding matches of misspellings and similar words) based on the Levenshtein Distance, and range searches (similar to Google’s numrange search). To learn more, here is a page that lists all of Lucene’s search functionality.

Conclusion

Now that you know that LinkedIn uses Lucene as their text search engine and you’ve seen all of the powerful search functionality Lucene has to offer – wouldn’t you like to be able to use wildcard searching, proximity search, term weighting, and fuzzy search when searching LinkedIn? I know I do! Those features can make a HUGE difference in the relevance of search results.

I’m still trying to figure out why LinkedIn doesn’t offer users all of Lucene’s search functionality as they’ve been using Lucene as their text search engine for at least 7 months now.

I’ve tried to communicate my search improvement suggestions to LinkedIn a couple of different ways. In June I sent message to Esteban Kozak – Senior Product Manager overseeing search at LinkedIn - via LinkedIn (of course) that detailed all of my suggestions for improving LinkedIn’s search functionality, including wildcard search, proximity search, and term weighting – and I haven’t received a response.

I also caught William Uranga Tweeting from a LinkedIn customer advisory session last week, so I DM’d him and let him know I had a list of search recommendations and he kindly let me send them to him via email so he could share them during the session at LinkedIn. William wrote a post about his customer advisory session experience at LinkedIn – you can read it here.

We can only hope that sometime in the near future LinkedIn taps into the awesome search power of Lucene, enabling users to take control of search relevance and tap into semantic search. I know I’ve got my fingers crossed!

Update!

Esteban Kozak replied to my message with this helpful response:

1- Prefix matching: we are currently evaluating the release of prefix matching for names in order to enable a quick way to navigate your contacts from the mobile application. Prefix matching for free text queries is very expensive because the query needs to be translated into a huge OR statement in the back end. There are better ways to solve this problem more elegantly. We are investigating alternative approaches like stemming, automatic expansion at query time and other techniques to ensure good recall.

2- Proximity search / Term weighting: These two are much easier to open up and will be available shortly.

Also – be sure not to miss LinkedIn Principle Search Engineer Jake Mannix’s thorough and detailed comments below.

It appears we have much to look forward to with regard to LinkedIn search functionality!