Our software runs in desktop applications, multi-tenant SaaS architectures, global server farms, and everything in between. It is fully multi-threaded and cost-effective to scale up and out.
The Sovren Document Converter converts resumes from their native formats to a straight ASCII text format that is usable by the Sovren Resume/CV Parser. The Sovren Document Converter does not require any of the authoring software (Microsoft Word, Adobe PDF, etc.), as it is completely self-contained. The Converter can convert dozens of resumes per second.
In addition to converting documents to text, the Sovren Document Converter can convert Microsoft Word documents into formatted HTML. This HTML version looks like the original Word document, but is actually safe and lightweight HTML (virus free!).
The Sovren Resume/CV Parser is a component that performs the following functions:
1. Parses converted resumes to extract hundreds of pieces of important data.
2. Outputs that data in an open industry-standard HR-XML Resume format, or to text, or to CSV.
Please see the Sovren Resume/CV Parser Overview Whitepaper
The earliest predecessor of the Sovren Resume/CV Parser was built in 1996. Sovren is the oldest surviving parsing technology and we were the first company to be certified by HR-XML.org on the Resume 2.x standard.
The Sovren Resume/CV Parser powers the two largest career sites on the Internet. You can't get more scalable than that. On modern server hardware, the Sovren Resume/CV Parser generally completes parsing in under 1 second (regardless of language or region). Higher quality resumes will parse faster than low quality resumes. Every single resume section type is configurable on-the-fly to be turned "on" or "off" as desired.
Based on a real-world customer experience in May 2011, batch processing 2.5 million resumes on a dual Xeon X5690 3.46 GHz yielded an average throughput of 35 resumes per second. That's 126,000 per hour, 3 million per day, 1.1 billion per year -- on one multi-core server running the default configuration of our BulkParser sample application.
Just as importantly, the parser easily scales to the low-end. Many customers add the parser to their existing application server (which is important for ISVs), and some even embed the parser directly into their desktop client applications.
Yes. The parser ships with a built-in skills taxonomy covering approximately 50 top level categories (Legal, Retail, Accounting and Finance, Medical, HR, IT, etc.); approximately 500+ sub-categories; and 25,000+ skills, some of which are further grouped into synonym hierarchies. More importantly the Sovren Resume/CV Parser will tell you where it found each skill (such as in work history, job objective, qualification summary etc.). Each skill or competency as they are tagged in HR-XML will also be tagged with the years of experience and when that skill was last used as identified with work experience. This is a key piece of information in beginning to understand who a candidate is, who she was, and what she may be qualified to do from a career perspective.
No problem. The parser allows you to create an unlimited number of custom skills taxonomies, and you can specify which taxonomies that you want the parser to use on a per-resume-parsed basis. You can use our skills list, add to our skills list, or build your own skills lists.
We can supply over a dozen ready-to-run, fully commented sample applications, which are all included with the license to the parser. Clients are free to modify and use the source code to these applications. Applications include:
The Sovren Document Converter is a.NET Assembly and/or an ActiveX control that converts resumes from their native formats to a straight ASCII text format that is usable by the Sovren Resume/CV Parser. This is the first step to normalize the data for the parser. The Sovren Document Converter does not require any of the authoring software (Microsoft Word, Adobe PDF, etc.), as it is completely self-contained. The Converter can convert as many as 100 resumes per second (YMMV).
The Sovren Document Converter, which is included with the Sovren Resume/CV Parser, provides lightning-fast conversion of essentially all commercially used recruitment document formats including Microsoft Word (including DOCX), Adobe PDF, Corel WordPerfect, HTML, MS Office HTML, Open Office (ODF), Rich Text (RTF), and various other formats.
For the parser, integration only requires the installation of two files - one for the Sovren Document Converter and one for the Sovren Resume/CV Parser. Click-and-go, zero-code, ready-to-run solutions are available at no extra charge. You can be running your own instance of the SovrenConvertAndParse Web Service in about 20 minutes. Installation of the optional Sovren Semantic Matching Engine is far more involved, and requires substantial integration after installation.
Yes, and Yes. The Sovren Resume/CV Parser is written completely in C# using the Microsoft .NET Framework. However, the Parser integrates seamlessly into COM-compatible environments such as Delphi 6-7 and Visual Basic 6. For clients who wish to integrate into COM environments, we supply the Parser as a late-bound COM control, and we supply the corresponding Type Library.
Integration with non-Windows environments is extremely simple and can be accomplished with any of our zero-code deployment options. However, as a Microsoft Windows .NET application, the Parser itself must be run on a Windows OS machine. About half of our customers primarily run non-Windows environments. About 50% of our customers are non-Windows shops who integrate the Parser via running an instance of the supplied SovrenConvertAndParse Web Service
YES. The Sovren Resume/CV Parser is a native C#/.NET component and the Sovren Document Converter is a .NET assembly (with optional additional Converter functionality provided through traditional Windows DLLs).
Yes. The Parser can run multi-threaded or single-threaded. In IIS webserver applications, IIS will automatically manage the thread pool for you since the parser is a native .NET component.
Yes, if approved, and at substantial additional cost.
Only the free runtime is required. The Parser runs on the .NET Framework v2.0 or higher runtime, which ships with MS Windows.