Expansion Methods for Job-Candidate Matching Amidst Unreliable and Sparse Data

We address the problem of matching jobs with workers when information about both elements is incomplete and in some cases inaccurate. Such a situation occurs, for example, when profile information is generated from recorded audio, rather than typed or written sources. We present various methods of dealing with such post-processed voice information and show how it compares to human generated matches over the same data. Our analysis includes both SQL- and ontological-based methods that provide higher recall over a sparse data. A probabilistic weighted ontology model is proposed that enables assignment of realistic weights to different attributes and considers probabilistic conversion of audio to text. The evaluation is performed on real-life data from 1,100 candidates and 48 jobs spanning more than 3,000 vacancies.

[1]  Karthik Visweswariah,et al.  PROSPECT: a system for screening candidates for recruitment , 2010, CIKM.

[2]  George A. Miller,et al.  WordNet: A Lexical Database for English , 1995, HLT.

[3]  Olivier Chapelle,et al.  Expected reciprocal rank for graded relevance , 2009, CIKM.

[4]  Balakrishnan Chandrasekaran,et al.  What are ontologies, and why do we need them? , 1999, IEEE Intell. Syst..

[5]  Gilad Mishne,et al.  Automatic analysis of call-center conversations , 2005, CIKM '05.

[6]  Dipanjan Chakraborty,et al.  VOISERV: Creation and Delivery of Converged Services through Voice for Emerging Economies , 2007, 2007 IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks.

[7]  David Carmel,et al.  Spoken document retrieval from call-center conversations , 2006, SIGIR.

[8]  Ellen M. Voorhees,et al.  The TREC Spoken Document Retrieval Track: A Success Story , 2000, TREC.

[9]  Steffen Staab,et al.  Gimme' the context: context-driven automatic semantic annotation with C-PANKOW , 2005, WWW '05.

[10]  Daniel S. Weld,et al.  Information extraction from Wikipedia: moving down the long tail , 2008, KDD.

[11]  Tim Weitzel,et al.  Decision support for team staffing: An automated relational recommendation approach , 2008, Decis. Support Syst..

[12]  Kalina Bontcheva,et al.  Ontology-Based Information Extraction for Business Intelligence , 2007, ISWC/ASWC.

[13]  Christie I. Ezeife,et al.  Using domain ontology for semantic web usage mining and next page prediction , 2009, CIKM.

[14]  Kun Yu,et al.  Resume Information Extraction with Cascaded Hybrid Model , 2005, ACL.

[15]  George A. Miller,et al.  Using Corpus Statistics and WordNet Relations for Sense Identification , 1998, CL.

[16]  Saurabh Srivastava,et al.  Designing a voice-based employment exchange for rural India , 2012, ICTD '12.

[17]  Arnaud Sahuguet,et al.  An audio indexing system for election video material , 2009, 2009 IEEE International Conference on Acoustics, Speech and Signal Processing.

[18]  Peng Yu,et al.  Vocabulary-independent indexing of spontaneous speech , 2005, IEEE Transactions on Speech and Audio Processing.

[19]  Alex Acero,et al.  Position Specific Posterior Lattices for Indexing Speech , 2005, ACL.

[20]  Tom E. Bishop,et al.  Blind Image Restoration Using a Block-Stationary Signal Model , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[21]  Hwee Tou Ng,et al.  A lattice-based approach to query-by-example spoken document retrieval , 2008, SIGIR '08.

[22]  Amit Singhal,et al.  Document expansion for speech retrieval , 1999, SIGIR '99.

[23]  Tobias Keim,et al.  Extending the Applicability of Recommender Systems: A Multilayer Framework for Matching Human Resources , 2007, 2007 40th Annual Hawaii International Conference on System Sciences (HICSS'07).