Artificial intelligence-based resource selection
A method includes retrieving information regarding a candidate from a plurality of sources, and analyzing the information regarding the candidate using one or more machine learning techniques. A plurality of questions for the candidate are generated based on the analysis. The method further includes receiving and analyzing a plurality of natural language responses to the plurality of questions from the candidate, and computing a plurality of confidence scores for the plurality of natural language responses using the one or more machine learning techniques. The plurality of questions and the plurality of confidence scores are provided to a user via a user interface.
1. An apparatus comprising:
at least one processing platform comprising a plurality of processing devices;
the at least one processing platform being configured:
to retrieve information regarding a candidate from a plurality of sources;
to analyze the information regarding the candidate using one or more machine learning models;
to generate a plurality of questions for the candidate based on the analysis;
to receive and analyze a plurality of natural language responses to the plurality of questions from the candidate;
to compute a plurality of confidence scores for the plurality of natural language responses using the one or more machine learning models; and
to provide the plurality of questions and the plurality of confidence scores to a user via a user interface;
wherein generating a plurality of questions for the candidate based on the analysis comprises:
generating the plurality of questions for the candidate based on data from a knowledge base, wherein the data from the knowledge base comprises data identifying one or more expertise areas of the user;
wherein receiving and analyzing includes utilizing the responses to the plurality of questions as input to train the one or more machine learning models to enable identification of a baseline speech pattern for the candidate;
wherein computing the plurality of confidence scores includes (i) identifying one or more other speech patterns for the candidate different from the baseline speech pattern, (ii) determining consistencies between the responses to the plurality of questions and the information regarding the candidate from the plurality of sources, and (iii) comparing the responses to the plurality of questions from the candidate to the same or similar responses given by one or more other candidates;
wherein at least one of the confidence scores of the plurality of confidence scores is representative of the one or more other speech patterns; and
wherein the at least one processing platform is further configured to:
generate at least a second plurality of questions for one or more additional candidates; and
dynamically update the knowledge base with the at least a second plurality of questions and answers to the at least a second plurality of questions provided by the one or more additional candidates.
2. The apparatus of claim 1 wherein the at least one processing platform is further configured to use natural language processing techniques to analyze the plurality of natural language responses.
3. The apparatus of claim 1 wherein, in retrieving the information regarding the candidate from the plurality of sources, the at least one processing platform is configured to use one or more network crawling techniques to extract the information from one or more of the plurality of sources.
4. The apparatus of claim 3 wherein the one or more of plurality of sources comprise at least one of a social media platform, online publications, webpages and online databases.
5. The apparatus of claim 1 wherein the plurality of sources comprises at least one of an uploaded resume and an uploaded curriculum vitae of the candidate.
6. The apparatus of claim 5 wherein, in analyzing the information regarding the candidate, the at least one processing platform is configured to cross-reference experience descriptions from different entities in at least one of the uploaded resume and the uploaded curriculum vitae.
7. The apparatus of claim 1 wherein the one or more machine learning models comprise a duo-directional attentive memory network.
8. The apparatus of claim 1 wherein the data from the knowledge base comprises a plurality of flags corresponding to a plurality of topics to avoid for the plurality of questions.
9. The apparatus of claim 1 wherein the at least one processing platform is further configured to generate one or more additional questions for the candidate based at least on one or more of the plurality of natural language responses.
10. The apparatus of claim 1 wherein the at least one processing platform is further configured to modify one or more of the plurality of questions based at least on one or more of the plurality of natural language responses.
11. The apparatus of claim 1 wherein computing the plurality of confidence scores for the plurality of natural language responses is performed in real-time.
12. The apparatus of claim 1 wherein the at least one processing platform is further configured to rank the plurality of confidence scores for the plurality of natural language responses.
13. A method comprising:
retrieving information regarding a candidate from a plurality of sources;
analyzing the information regarding the candidate using one or more machine learning models;
generating a plurality of questions for the candidate based on the analysis;
receiving and analyzing a plurality of natural language responses to the plurality of questions from the candidate;
computing a plurality of confidence scores for the plurality of natural language responses using the one or more machine learning models; and
providing the plurality of questions and the plurality of confidence scores to a user via a user interface;
wherein generating a plurality of questions for the candidate based on the analysis comprises:
generating the plurality of questions for the candidate based on data from a knowledge base, wherein the data from the knowledge base comprises data identifying one or more expertise areas of the user;
wherein receiving and analyzing includes utilizing the responses to the plurality of questions as input to train the one or more machine learning models to enable identification of a baseline speech pattern for the candidate;
wherein computing the plurality of confidence scores includes (i) identifying one or more other speech patterns for the candidate different from the baseline speech pattern, (ii) determining consistencies between the responses to the plurality of questions and the information regarding the candidate from the plurality of sources, and (iii) comparing the responses to the plurality of questions from the candidate to the same or similar responses given by one or more other candidates;
wherein at least one of the confidence scores of the plurality of confidence scores is representative of the one or more other speech patterns;
wherein the method further comprises:
generating at least a second plurality of questions for one or more additional candidates; and
dynamically updating the knowledge base with the at least a second plurality of questions and answers to the at least a second plurality of questions provided by the one or more additional candidates; and
wherein the method is performed by at least one processing platform comprising at least one processing device comprising a processor coupled to a memory.
14. The method of claim 13 further comprising generating one or more additional questions for the candidate based at least on one or more of the plurality of natural language responses.
15. The method according to claim 13 wherein the data from the knowledge base comprises a plurality of flags corresponding to a plurality of topics to avoid for the plurality of questions.
16. The method according to claim 13 wherein the one or more machine learning models comprise a duo-directional attentive memory network.
17. The method according to claim 13 further comprising modifying one or more of the plurality of questions based at least on one or more of the plurality of natural language responses.
18. A computer program product comprising a non-transitory processor-readable storage medium having stored therein program code of one or more software programs, wherein the program code when executed by at least one processing platform causes the at least one processing platform:
to retrieve information regarding a candidate from a plurality of sources;
to analyze the information regarding the candidate using one or more machine learning models;
to generate a plurality of questions for the candidate based on the analysis;
to receive and analyze a plurality of natural language responses to the plurality of questions from the candidate;
to compute a plurality of confidence scores for the plurality of natural language responses using the one or more machine learning models; and
to provide the plurality of questions and the plurality of confidence scores to a user via a user interface;
wherein generating a plurality of questions for the candidate based on the analysis comprises:
generating the plurality of questions for the candidate based on data from a knowledge base, wherein the data from the knowledge base comprises data identifying one or more expertise areas of the user;
wherein receiving and analyzing includes utilizing the responses to the plurality of questions as input to train the one or more machine learning models to enable identification of a baseline speech pattern for the candidate;
wherein computing the plurality of confidence scores includes (i) identifying one or more other speech patterns for the candidate different from the baseline speech pattern, (ii) determining consistencies between the responses to the plurality of questions and the information regarding the candidate from the plurality of sources, and (iii) comparing the responses to the plurality of questions from the candidate to the same or similar responses given by one or more other candidates;
wherein at least one of the confidence scores of the plurality of confidence scores is representative of the one or more other speech patterns; and
wherein the program code when executed by the at least one processing platform further causes the at least one processing platform:
to generate at least a second plurality of questions for one or more additional candidates; and
to dynamically update the knowledge base with the at least a second plurality of questions and answers to the at least a second plurality of questions provided by the one or more additional candidates.
19. The computer program product according to claim 18 wherein the data from the knowledge base comprises a plurality of flags corresponding to a plurality of topics to avoid for the plurality of questions.
20. The computer program product according to claim 18 wherein the one or more machine learning models comprise a duo-directional attentive memory network.