Lens/Xray™

Lens/Xray™ offers unparalleled parsing capabilities, extracting data from free text resumes and job postings and providing output in the form of HR-XML standard documents. Most parsing applications use semantic and rules-based systems, but Lens/Xray is different. Leveraging a field of artificial intelligence known as Statistical Natural Language Processing, Lens/Xray emulates the human brain by using context to categorize information based on what is being described, evaluating each word independently with reference both to meaning and contextual cues.

As a result, Lens/Xray delivers the highest levels of accuracy in the industry. Because it relies primarily on context and observed patterns of writing rather than fixed rules, it is not confounded by non-standard formats or non-sequential information and is particularly adept when it comes to extracting information from complex, unstructured fields like work history. In addition, Lens/Xray normalizes and infers far more data than other solutions, including standardized geocodes, industry codes, functional codes, educational levels, academic major codes, and dates and years of experience. Skills are rolled up into a standardized, hierarchical, and customizable dictionary that also includes context, duration, and recentness of use for each skill.

Lens/Xray accepts documents in most major document formats, including Word, PDF, WordPerfect, HTML, RTF, and plain text. Lens/Xray also accepts forms-driven input from candidates who lack a resume or when customer requirements dictate completion of a standard application form.

Lens/Xray is language independent, which means it easily handles foreign languages, including French, Dutch, German, Chinese, and other multi-bit languages.

• Lens/Xray™ leverages patented neural network technology to identify candidate data directly from resumes and job postings with the highest degree of precision. Using a set of techniques called Statistical Natural Language Processing (SNLP), our artificial intelligence engine is able to read and understand resumes by making probabilistic determinations based on linguistic context and inference.
• The core of Lens/Xray is a mathematical, statistical model which uses word meaning, context, sentence structure, and surroundings to determine form, subject, and meaning. It is trained on thousands of resumes and job descriptions previously annotated by live experts. Over many trials, it has learned through case-by-case experience to mimic the behavior of its human annotators, to the point where it sometimes outperforms them.
• Lens/Xray is a self-adaptive machine learning technology. It starts with substantial intelligence out of the box but continues to learn as it goes along, identifying and analyzing emerging trends in order to update automatically its underlying tables and dictionaries. Burning Glass scientists also deliver updates to each of our models multiple times each year to expand and refresh all dictionaries, data tables, and coding schemas without the need for client maintenance.
• The algorithms used by Lens/Xray are powered by data contained in Burning Glass’s Knowledge Mine™, a unique repository of historical career data extracted from millions of hiring decisions. This is important because no two resumes are identical, either in format or content. To keep Lens/Xray in tune with emerging trends, the Knowledge Mine is updated with new resumes on a yearly basis. Models are also geography specific, so that every country has its own model which captures details about candidate skills, knowledge, experience, and career transition that are characteristic of the local employment market.
• Extensible Mark Up Language (XML) is a human-readable, industry-standard format for diverse information. It relies on inserted markers (tags) which describe nearby text. Typical tags inserted by Lens/Xray include “the beginning of the title of the second job from the first employer described in the resume’s experience section”. Lens/Xray tags also contain information that it has deduced, including address latitude and longitude, SICs for each employer, standardized job title for each job, level of each academic degree, and major area of study. All of these are automatically determined by Lens/Xray based on its analysis of resume data.
• Lens/Xray output may be coupled with standard style sheets to specify and render resumes in any standard format you desire, from summarizations to industry-standard HR-XML format.

• Real Savings – Lens/Xray™ eliminates the cost and drudgery of manual data entry, letting your staff focus on hiring instead of processing. Lens/Xray also delivers dramatic savings compared to outsourced resume processing services.
• Superior Candidate Experience – Job seekers simply drop their resumes and go. There are no forms to complete, making it easier to apply and increasing your applicant flow.
• Instantaneous Results – Because Lens/Xray processes resumes automatically, it eliminates the delays caused by application processing and accelerates your operations.
• Most Comprehensive Data Capture – Lens/Xray doesn’t stop with the basics. It extracts hundreds of highly relevant candidate facts and stores them for easy and relevant search and retrieval.
• Highest Precision in the Industry – Lens/Xray was designed exclusively to read and extract information from resumes and job descriptions. As a result, it is far more accurate than general parsing engines in identifying candidate data, making it the most reliable resume parsing technology available. Our technologies employ sophisticated and proprietary Statistical Natural Language Processing techniques that are on the cutting edge of academic research in artificial intelligence.
• Drives More Relevant Searching – Superior data extraction provides the basis for more targeted searching. Use parsed output to power contextual search or as the basis for deploying intelligent search and match modules.
• Easy In, Easy Out – Lens/Xray processes resumes in most major document formats (including Microsoft Word, RTF, plain text, ASCII, HTML, WordPerfect, PDF) and produces output in XML or mapped to your specific database configuration.
• Faster Resume Review – Use Lens/Xray to reformat candidate resumes in a standardized template for quick, side-by-side review. Because Lens/Xray identifies and extracts all of the data from the resume, you can present it in the format of your choosing. This makes it much easier for users to find the information they need because it will always be in the same place.
• Enhance Brand Identity – Recruitment agencies and job boards are justifiably proud of the candidates they put forward to potential employers. By reformatting candidate resumes in a custom template, you can make your candidates’ sourcing more easily identifiable.

The technology that Lens/Xray™ uses to read, understand, and catalog information directly from resumes and job postings is truly state-of-the-art. Unlike other engines which rely on fixed rules and predefined templates, Burning Glass has developed patented technology for data extraction that leverages a field of artificial intelligence known as Statistical Natural Language Processing (SNLP). Using SNLP technology, Lens/Xray learns from the way the millions of past resumes and job descriptions it has processed were written and formatted. Rather than looking for information to be in specific sequences or formats, the system uses context to categorize information based on what it describes, evaluating each word independently based both on meaning and contextual cues. For example, it knows the difference between someone who has worked for the University of Iowa and someone who has attended the University of Iowa or between someone who has prepared budgets and someone who has worked at Budget Rent a Car. Others use rules, fixed formulas, and pre-defined templates to drive data extraction, which means that they can be easily confounded when resumes are written in non-standard formats or when information sequence is jumbled in a table conversion.

As a result, Lens/Xray delivers the highest levels of accuracy in the industry. Because it relies primarily on contextual cues and observed patterns of resume writing (as described above) rather than fixed rules, the engine is particularly adept when it comes to discerning information from complex fields like work history – which can be quite unstructured.

In addition, Lens/Xray goes far beyond other solutions when normalizing the data that it extracts. This includes automated inference of standardized geocodes, industry codes, functional codes, educational levels, academic major codes, and normalization of dates and years of experience. Skills are rolled up to a standardized, hierarchical, and customizable dictionary of skills that also includes a full contextual summary including context, duration, and recentness of use. All coding schemes and skills dictionaries are fully customizable.

The implications of this are profound. By coding and normalizing a wide array of information both from resumes and jobs, Lens/Xray delivers the industry’s most powerful structured search capabilities. Employers can screen applicants based upon information included within their resumes and job seekers can search for jobs based upon information read directly from postings. Enabled search features include easy-to-use range-bound (e.g., "find jobs with salary between $30,000 and $40,000 per annum") and contextual search mechanisms (e.g., find candidates who have worked for Microsoft, not those who have used Microsoft Office). Additional structured search criteria include experience level, degree level, specific degrees or certifications, salary, industry code, functional code, and distance. This largely removes the ambiguity and hit-or-miss nature of searching for such parameters via keywords. For example, employers searching for college graduates using a keyword search engine could specify "college", "bachelor", "BA", "BS", "B.S.", and "B.A.", and still miss candidates with degrees such as "LLB" or "BN" or "B.Sc", among many others. By contrast, Lens/Xray normalizes all college credentials to a cataloged degree name and to a standardized degree level for easy search, retrieval, and reporting... even when the specific degree is unknown.