WO2021202407A1

WO2021202407A1 - Computer platform implementing many-to-many job marketplace

Info

Publication number: WO2021202407A1
Application number: PCT/US2021/024704
Authority: WO
Inventors: Ashutosh Garg; Yuet Ping Poon; Anthony HAHN
Original assignee: Eightfold AI Inc.
Priority date: 2020-03-30
Filing date: 2021-03-29
Publication date: 2021-10-07

Abstract

A system and method for a multi-level job marketplace include one or more processing devices to identify, based on one or more features, a first set of talent profiles from a pool of talent profiles supplied by a first information system of a first entity, combine the first set of talent profiles to generate a talent profile collector, execute a first neural network module to generate a first-level match score between the talent profile collector and a job profile collector provided by a second information system of a second entity, and determine whether to provide the first set of talent profiles to the second entity based on the first-level match score.

Description

Computer Platform Implementing Many-to-Many Job Marketplace

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Application No. 16/953,884 filed on November 20, 2020, which claims priority to U.S. Provisional Application 63/002,301 filed March 30, 2020. The above-mentioned applications are hereby incorporated by reference in their entireties.

TECHNICAL FIELD

[0002] The present disclosure relates to a computer platform implementing a job marketplace that facilitates and enables quick job matching at the organization level, namely between groups of job candidates with one or more job listings, and in particular to a computer platform implementing many-to-many job marketplace for matching combinations of multiple job candidates from multiple candidate supplying companies with multiple job openings of multiple hiring companies.

BACKGROUND

[0003] An organization may be composed of units of employees. The organization can be a company, a nonprofit organization, or a government agency. The units of employees can be a division, a department, a team, a group associated with a project, or any combination of employees grouped together for one or more common objectives or defined by one or more common features. The employees of the organization may include experienced and new employees. The new employees may include fresh recruits from schools with little or no prior work experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0004] The disclosure will be understood more fully from the detailed description given below and from the accompanying drawings of various embodiments of the disclosure. The drawings, however, should not be taken to limit the disclosure to the specific embodiments, but are for explanation and understanding only.

[0005] FIG. 1 illustrates a computing system for providing a job marketplace for exchanging the information of a unit of employees between candidate suppliers and job providers according to an implementation of the disclosure.

[0006] FIG. 2 illustrates computing system for providing a job marketplace for multi level job matching according to an implementation of the disclosure.

[0007] FIG. 3 illustrates a machine learning module according to an implementation of the disclosure.

[0008] FIG. 4 illustrates workflows for the candidate supplier, the at-risk employees, and the job provider according to an implementation of the disclosure.

[0009] FIG. 5 illustrates a process for an at-risk employee to use the job marketplace according to an implementation of the disclosure.

[0010] FIG. 6 illustrates a flowchart of a method for implementing a job marketplace according to an implementation of the disclosure.

[0011] FIG. 7 depicts a block diagram of a computer system operating in accordance with one or more aspects of the present disclosure.

DETAILED DESCRIPTION

[0012] An organization during extraordinary economic situations may need to furlough or even lay off a large number of employees to relieve the economic stress on the organization (referred to hereinafter as the “candidate supplier”). The employees can be permanent employees, termed employees, or contractors. The layoff may target employees cross the board or units of employees, where a unit can be a division, a department, a team, or a group associated with a project. For example, a pandemic breakout can severely impact the operations of retailers, transportation industries, and entertainment venues. The quick and steep deterioration of macro and micro economic environments may force a candidate supplier to furlough and/or lay off employees in mass. The candidate supplier may have a target number of at-risk employees to be placed in furlough and layoff programs. Because the at-risk employees are not expecting the sudden loss of their positions with the organization, the candidate supplier may be motivated to assist the job placements of these employees. On the other hand, other organizations (referred to as the “job providers”) in the same situation may experience increased business demands and need to hire employees in mass to fulfill a lot of job openings quickly. For example, during the same pandemic, the online retailers may experience sudden increased demands for food or household item deliveries.

[0013] Traditional job replacement processes would include the following steps. The candidate supplier may furlough or lay off the employees and provide the at-risk employees with outplacement assistance programs which may include training classes and job leads. Each of the employees may prepare his or her resume with or without their employer’ s assistance, and submit the resume to different potential hiring employers based on the employee’ s preference. The hiring employers may assess the talent based on the received resume, and decide whether to initiate a hiring process that may include interviews and hiring decisions. This traditional job replacement process, however, is slow and inefficient in placing a large number of employees during these extraordinary situations. The traditional job replacement process does not match at a level beyond individual job matching. Instead, the traditional job replacement process requires each job candidate to individually search and apply for a job opening, resulting an inefficient job matching process with respect to an organization when the organization needs to outplace a large number of candidates in a short time period. In such a situation, the candidate supplier as the current employer may desire to minimize the average time for the employees to transition to their next jobs and to maximize the opportunities for all of the at-risk employees as a whole. The traditional processes do not scale to exchange jobs between organizations at massive numbers in a short time period (e.g., in a matter of days).

[0014] To overcome the above-identified and other technical problems in the practical application of assisting a large number of at-risk employees to transition into next jobs made available by other employers, implementations of the disclosure provide a computer platform that may facilitate the job matching among organizations to a level beyond individual job matching. This match at a level beyond just an individual (e.g., at an organization level, at a team level, or at a project level) may also allow the outplacing outplace the at-risk employees with business competitors. The computer platform may assess the matching between the candidate supplier organization and each of hiring organizations at more than the individual level, and determine one or more job providers based on the assessed matching. Implementing an intelligent job marketplace that may identify job exchanges based on certain criteria regarding underlying relationship and contents and permission setting of information exchanges between the job providers and candidate supplier. For example, the system may allow a job provider to specify that a set of teams or projects at the organization cannot be matched to certain teams or projects from a particular set of candidate supplying organizations; conversely it may provide affirmative preferences to hire from certain candidate supplying organizations or to hire from individuals with a history of certain teams or projects. The candidate supplier may also have similar outgoing permission settings. In terms of information exchange, for instance, the candidate supplier may allow certain employment information to be passed to the job providers such as background report date, background check status, or whether it will continue to supply benefits or not within the bound of privacy laws and regulations. The hiring company may pass information back such as willingness to charge back costs of employees to the candidate supplier.

[0015] Implementations of the disclosure may include a computer system that may implement a multi-level job placement platform. The computer system may include a memory device, and one or more processing devices, communicatively connected to the memory device. The one or more processors may identify, based on one or more features, a first set of talent profiles supplied by a first information system of a first entity, combine the first set of talent profiles to generate a talent profile collector, execute a first neural network module to generate a first-level match score between the talent profile collector and a job profile collector provided by a second information system of a second entity, and determine whether to provide the first set of talent profiles to the second entity based on the first- level match score. The talent profile collector can be a data object that includes information of multiple talent profiles that are associated with personnel of a candidate supplier or a unit of the candidate supplier. The job profile collector can be a data object that may include one or more job profiles associated with job provider, where the one or more job profiles may define the needs of a unit of the job provider. Further, the one or more processors may responsive to determining to provide the first set of talent profiles to the second entity based on the first-level match score, execute a second neural network module to generate a respective second-level match score between each of the first set of talent profiles and each job profile in the job profile collector, and present, in a first user interface, the second-level match scores in a ranked order.

[0016] FIG. 1 illustrates a computing system 100 for providing a job marketplace 108 for exchanging the information of a unit of employees between candidate suppliers and job providers according to an implementation of the disclosure. Computing system 100 can be a standalone computer or a networked computing resource implemented in a computing cloud. Referring to FIG. 1, computing system 100 may include one or more processing devices 102, a storage device 104, and an interface device 106, where the storage device 104 and the interface device 106 are communicatively coupled to processing devices 102.

[0017] A processing device 102 can be a hardware processor such as a central processing unit (CPU), a graphic processing unit (GPU), or an accelerator circuit. Interface device 106 can be a display such as a touch screen of a desktop, laptop, or smart phone. Storage device 104 can be a memory device, a hard disc, or a cloud storage connected to processing device 102 through a network interface card (not shown).

[0018] Processing device 102 can be a programmable device that may be programmed to implement a graphical user interface presented on interface device 106. The interface device may include a display screen for presenting textual and/or graphic information. Graphical user interface (“GUI”) allows a user using an input device (e.g., a keyboard, a mouse, and/or a touch screen) to interact with graphic representations (e.g., icons) presented on GUI.

[0019] Computing system 100 may be connected to other information systems 110, 114A, 114B through network (not shown). These information systems can be human resource management (HRM) systems that are associated with different organizations. The HRM systems can track external/intemal candidate information in the pre-hiring phase (e.g., using an applicant track system (ATS)), or track employee information after they are hired (e.g., using an HR information system (HRIS)). Thus, these information systems may include databases that contain information relating to candidates and current employees. Implementations of the disclosure provide a job marketplace 108 implemented on computing system 100 that may facilitate the HR information exchange between the HRM systems of candidate suppliers and the HRM systems of job providers.

[0020] Job marketplace 108 may serve as an intermediate system among different organizations with trust from the candidate suppliers and job providers, thus isolating the HRM system of a candidate supplier from direct access by the HRM system of a job provider. This implementation has the advantage of protecting the information stored in the HRM system of the candidate supplier from unauthorized accesses, thus protecting the privacy of these candidates. In addition to isolating the disparate HRM systems that belong to different organizations, job marketplace 108 can be a platform that provides intelligent, multi-level matching service that may match a large quantity of at-risk employees (or potential job candidates) of a candidate supplier to a group of job openings of a job provider. The job marketplace 108 may include trained machine learning module (e.g., trained neural network module) to perform the intelligent selection from the candidates to generate a selected group of candidates that may match to the group of the job openings of the job provider. In one implementation, instead of trying to match one candidate with one job opening, job marketplace 108 may perform multi-level matching in which a first-level of matching may include a many-to-many matching that identify, using the trained machine learning module, a group of candidates matched to a group of job openings. In one implementation, the group of candidates may be represented by a talent profile collector including the talent profiles of these candidates, and the group of job openings may be represented by a job profile collector. This many-to-many matching can significantly reduce the time to identify a large number of candidates for a large number of job openings, which is particularly useful in situations where an organization needs to quickly outplace a large number of employees and another organization needs to screen and hire a large number of employees. Subsequent to the first-level of matching, job marketplace 108 may further calculate a second-level match score for each candidate from the selected group of employees with respect to a job opening from the group of job openings, thus providing a ranked list of candidates for each job opening. Using this multiple level matching, job marketplace 108 performs quick matching between the large quantity of at-risk employees of the candidate supplier and the job openings of the job provider.

[0021] Through the multi-level matching process, job marketplace 108 implemented on computing system 100 may obtain relevant information (or filtered information) of candidates that may meet the requests of the job provider. Computing system 100 may notify the matching to candidate supplier and/or the job provider to obtain their permission to exchange the employee information. Upon their approvals, job marketplace 108 may provide all or a filtered version of the identified candidates to the job provider. In this way, these organizations may exchange employee information or filtered employee information using computing system 100 to facilitate the placement of a large quantity of employees from one organization to another.

[0022] Referring to FIG. 1, information system 110 can be a HRM system that belongs to a candidate supplier which is an organization, due to different reasons (e.g., economic distress), plans to lay off or furlough a large number of employees. Information system 110 may include a database that stores the talent profiles 112 of the at-risk employees. The talent profiles 112 can be a data object that contains data points related to an employee. In some implementations, the talent profile 112 may include a job title currently held by the employee and job titles previously held by the employee, teams to which the employee previously belonged and currently belongs or projects on which the employee previously worked and currently works, the technical or non-technical skills possessed by the employee for performing the job held by the employee, and the location (e.g., city and state) of the employee. The talent profile 112 may further include the employee’s education background information including schools he or she has attended, fields of study, and degrees obtained. The talent profile 112 may further include other professional information of the employee such as professional certifications the employee has obtained, achievement awards, professional publications, and technical contributions to public forum (e.g., open source code contributions). In addition to these fact-based data points, the talent profile may also be enriched to include derived information relating to the employee. For example, talent profile 112 may include predicted values that indicate the likely career path through the organization if the employee stays with the organization for a certain period of time. The career path may indicate the potential of the employee with the organization. The predicted items may include characteristics of the worker calculated from the information pertaining to the worker. These characters may include the effort level, the leadership, the velocity of skill improvement, the velocity of promotions etc. Information system 110 may also include other information relating to the employee such as whether the employee is at risk for a permanent layoff or a temporary furlough, and if it is the furlough, how long the furlough is. For example, a furlough worker may be available for other companies for a period of three months, six months etc., and a laid-off worker may be indicated as available for permanent positions. All of these data points relating to the employee may be stored in information system 110 in the form of a talent profile 112 that can be used by job marketplace 108 to evaluate job opportunities matching to the talent profile. [0023] Computing system 100 may be connected to information systems 114A, 114B that may each belong to a job provider that may be in the market to hire employees to fill urgent needs. Each information system 114A, 114B may include job profiles 116A,

116B that specify different aspects of the job openings. In one implementation, a job profile may include job requirements such as job titles, teams to which the hire belongs to, projects on which the hire works, job functions, required experiences, requisite education/degrees/certificates/licenses etc. The job profiles may also include desired personality traits of the candidates such as leadership attributes, social attributes, and altitudes. In addition to these express requirements that can be specified in a textual description, a job profile may also include the talent profiles of employees that had been hired for the same or similar positions and the talent profiles of candidates that the organization considered to hire. These talent profiles may contain information that can be uncovered in the machine learning module.

[0024] To perform the multi-level matching for the large quantity of talent profiles provided by information system 110, processing devices 102 in computing system 100 may execute job marketplace 108 to perform operations 118. At 120, processing devices 102 may identify, based on one or more features, a set of talent profiles from the talent profiles 112 received from information system 110. As discussed before, talent profiles 112 may represent those of at-risk employees of candidate supplier. The one or more features can be used to narrow down the pool of talent profiles 112 into a relevant set of talents. In one implementation, the one or more features can be one or more of a job title, a job identification code, a required job skill, a team identifier, a project identifier, a required education achievement (e.g., a degree, a professional certification) etc. Job marketplace 108 may identify a set of talent profiles that each is associated with at least one of the one or more features.

[0025] At 122, processing devices 102 may further combine the set of talent profiles, identified based on the one or more features, to generate a talent profile collector. The talent profile collector can be a container data object that includes all elements of the set of talent profiles. The talent profile collector may represent the fundamental and distinctive characteristics or qualities (referred to as the DNA) of the set of talent profiles. In one implementation, processing device 102 may generate the talent profile collector by placing all data items of the set of talent profiles into one container data object without any further processing. In another implementation, each talent profile may include a vector of category entries, where category entry can be a particular feature such as, for example, job titles, teams, projects, skills, experience, education, certifications, licenses etc. Processing devices 102 may combine data items of the set of talent profiles according to their category entries (i.e., data items that belong to one category are combined together). In yet another implementation, processing device 102 may combine the set of talent profiles by calculating a union of data items in the set of talent profiles. For example, for each category entry of text items, processing devices 102 may place unique words into the corresponding category entry in the talent profile collector; for each category of numerical data items, processing devices 102 may determine the minimum data range that covers all of the numerical data items and place the data range in the corresponding category entry in the talent profile collector.

[0026] At 124, processing devices 102 may execute a machine learning module to calculate a first- level match score between the talent profile collector and a job profile collector provided by information system 114A, 114B that belongs to a job provider.

The job profile collector can be a data object that contains one or more job profiles associated with the job provider. In one implementation, the job profile collector may define the needs of a team, a project, or a division of the job provider. The machine learning module can be an be a suitable statistical model or a deep neural network (DNN) that may be trained using training datasets.

[0027] The computing systemlOO may execute the job marketplace 108 for the set of talent profiles against all job openings, and rank the suitability of the jobs with respect to the at-risk employees based on the calculated first-level match scores. In similar way, the computer platform may determine a ranked list of job openings for each at-risk worker. In this way, the candidate supplier may provide each at-risk worker a customized list of job openings that is prepared based on the calculated match scores during outplacement. Since the list of job openings are determined by the machine learning model based on the talent profile and the job profile, the list of job openings may enhance the chance for the worker to be hired by a job provider, thus reducing the average time for the at-risk workers to transition into a next job.

[0028] FIG. 3 illustrates a machine learning module 200 according to an implementation of the disclosure. In one implementation, machine learning module 200 may be a deep neural network that may include multiple layers, in particular including an input layer for receiving data inputs, an output layer for generating outputs, and one or more hidden layers that each includes linear or non-linear computation elements (referred to as neurons) to perform the DNN computation propagated from the input layer to the output layer that may transform the data inputs to the outputs. Two adjacent layers may be connected by edges. Each of the edges may be associated with a parameter value (referred to as a synaptic weight value) that provide a scale factor to the output of a neuron in a prior layer as an input to one or more neurons in a subsequent layer. [0029] Referring to FIG. 3, machine learning module 200 may include an input layer including a first input 202A to receive a talent profile collector generated from the talent profiles of a group of employees and a second input 202B to receive a job profile collector containing jobs available to these employees. As discussed above, the talent profile is a data object including entries specifying different aspects of the employee.

The information relating to the employee may include aspects obtained from the HR database and may also include aspects obtained from external data sources such as professional web page, publications, and professional contributions to the public domains. Thus, the talent profile of the employee received at input 202A may include information beyond commonly available to an HR manager within the organization. The talent profile collector can be a combination of a set of talent profiles. Similarly, at discussed above, a job profile collector received at input 202B may include, but not limited to, titles and ranking levels of jobs, and skills required to perform the jobs, a minimum number of years of working experience. The machine learning module 200 may include an output layer including output 204 to produce a relevancy measurement, where the relevancy measurement is a parameter indicating how well the group of employees represented by the talent profile collector are matched to the job profile collector.

[0030] Responsive to receiving the talent profile collector of the set of employees at 202A supplied by information system 110 and the job profile collector at 202B provided by information system 116A, 116B, processing devices 102 may execute machine learning module 200 to calculate the relevancy measurement between the talent profile collector representing the set of talent profiles and the job profile collector, and further provide the relevancy measurement at output 204. In one implementation, the relevancy measurement may be served as a prediction indicator that may be used to predict which jobs are relevant to the set of employees. For example, if the relevancy measurement is greater than a threshold value, the jobs are relevant to the set of talent profiles. In another implementation, the relevancy measurement can be a correlation indicator that may be used to indicate how closely jobs are related to these at-risk employees. For example, if the relevancy measurement is greater than a threshold value, the jobs are closely related to the at-risk employees of the candidate supplier. In either case, the set of talent profiles are determined to be relevant to the jobs or not through machine learning module 200. FIG. 3 illustrates using machine learning module 200 to calculate the relevancy measurement between the talent profile collector and job profile collector. In another implementation, when there are multiple job providers to provide job profiles 116A, 116B to computing system 100, the machine learning module 200 may be applied to each pair of talent profile collector and job profiles to calculate the corresponding respective relevancy measurement to calculate the corresponding relevancy measurements. Processing devices 102 may determine, based on the relevancy measurements, the best jobs that match the talent profile collector.

[0031] Machine learning in this disclosure refers to methods implemented on hardware processing device that uses statistical techniques and/or artificial neural networks to give computer the ability to learn as the computer progressively improves performance on a specific task, from data without being explicitly programmed. The machine learning may use a parameterized model (referred to as “machine learning module”) that may be deployed using supervised learning/semi- supervised learning, unsupervised learning, or reinforced learning methods. Supervised/semi-supervised learning methods may train the machine learning modules using labeled training examples. To perform a task using supervised machine learning module, a computer may use examples (commonly referred to as “training data”) to test the machine learning module and to adjust parameters of the machine learning module based on a performance measurement (e.g., the error rate). The process to adjust the parameters of the machine learning module (commonly referred to as “train the machine learning module”) may generate a specific model that is to perform the practical task it is trained for. After training, the computer may receive new data inputs associated with the task and calculate, based on the trained machine learning module, an estimated output for the machine learning module that predicts an outcome for the task. Each training example may include input data and the corresponding desired output data, where the data can be in a suitable form such as a vector of numerical alphanumerical symbols.

[0032] The learning process of the machine learning module may be an iterative process. The process may include a forward propagation process to calculate an output based on the machine learning module and the input data fed into the machine learning module, and then calculate a difference between the desired output data and the calculated output data. The process may further include a backpropagation process to adjust parameters of the machine learning module based on the calculated difference.

[0033] In implementations of the disclosure, the training data may be constructed from historical data (e.g., prior hired employees or considered candidates for job openings). The relevancy measurement may be set based on a number of factors such as the percentage of candidates that are hired as employees for that job. A specific machine learning module 200 may be constructed through the training process using the train data set.

[0034] Referring to FIG. 1, the calculated first-level match score may be derived from the relevancy measurement between the talent profile collector and the job profile collector. The match score may be related to the probability of the at-risk employees associated with the set of talent profiles will be hired for the jobs. The first- level match score may represent a global likelihood of match between a group of candidates with a job profile collector representing one or more jobs.

[0035] At 126, processing devices 102 may determine, based on the first-level match score, whether to provide the first set of talent profiles to information system 114A,

114B. Responsive to determining that the first- level match score meets a pre-determined condition (e.g., equal to or above a threshold value) with respect to one or more job profiles belonging to a specific job provider, processing devices 102 may notify the information system (114A or 114B) of the job provider about the availability of the set of talent profiles from the candidate supplier and notify the candidate supplier about the availability of jobs for its at-risk employees. Processing devices 102 may further obtain permissions from the candidate supplier and the job provider to exchange information of the talent profiles and jobs through job marketplace 108. With the permissions both from the candidate supplier and the job provider, job marketplace 108 may facilitate the information exchange, thus achieving quick, batch job hunting for a large number of at- risk employees.

[0036] The operations 118 as shown in FIG. 1 may identify a match for a large number of candidates supplied by a first organization with jobs of a second organization to speed up the outplacement of at-risk employees. Thus, the first-level match score may represent a high-level matching at the organization level. When job marketplace 108 receives permissions from the candidate supplier and the job provider, job marketplace 108 may exchange the set of talent profiles and the job profiles between the candidate supplier and the job provider. In addition to the high-level matching at the organization level, implementations of the disclosure may provide a second-level matching at the level of individual candidate.

[0037] FIG. 2 illustrates computing system 100 for providing job marketplace 108 for multi-level job matching according to an implementation of the disclosure. As shown in FIG. 2, in addition to the first-level matching, job marketplace 108 may provide further a second-level, fine-grained matching at the individual candidate level. As shown in FIG. 2, at 128, processing devices 102 may further execute a second machine learning module to generate a respective second-level match score between each of the set of talent profiles and the at least one job profile in the job profile collector. The second machine learning module may have been trained using a second training data set. The second training data set may be constructed from the talent profiles of employees and their corresponding job profiles of the jobs held by these employees. The employees can be current employees and/or ex-employees of the job provider.

[0038] Each calculated second-level match score may indicate the relevancy measurement between each one of the set of talent profiles and a job profile. Thus, the matching of each candidate associated with the set of talent profiles may has a corresponding second-level match score.

[0039] At 130, processing devices 102 may present, in a user interface of user interface device 106, the second-level match scores in a ranked order. The ranked order may be in an order of candidates from high to low match scores, thus facilitating the job provider to review the candidates easily.

[0040] Implementations of the disclosure may provide workflows to the candidate supplier, the at-risk employees, and the job provider. FIG. 4 illustrates workflows 300 for the candidate supplier, the at-risk employees, and the job provider according to an implementation of the disclosure. Referring to FIG. 4, workflows 300 may be implemented in a computing system (e.g., computing system 100 as shown in FIG. 1). Workflows 300 may include a workflow 302 implemented for the candidate supplier which is an organization planning to furlough or lay off at-risk employees. Workflows 300 may further include a workflow 304 for each of the at-risk employees. Workflows 300 may further include a workflow 306 for the job provider which may intend to hire employees to fulfill job openings. Each workflow 302 - 304 may include a series of steps.

[0041] Workflow 302 for the candidate supplier may include step 308 to provide a land dashboard for the candidate supplier; step 310 for the candidate supplier to upload talent profiles of the at-risk employees; step 312 for the candidate supplier to contact the at- risk employees about using the job marketplace for outplacement; and step 314 for the candidate supplier to monitor and track the status of the at-risk employees’ job application.

[0042] Workflow 304 for the at-risk employees may include steps for interacting with workflow 302, these steps including step 316 to provide at-risk notice to the at-risk employee; step 318 to provide an opt- in page to the at-risk employee; responsive to opting-in by the at-risk employee, step 320 to provide a landing page to the at-risk employee where the at-risk employee may create a personal account with the job marketplace; step 322 to present enrollment questions and answers (e.g., personal identification, working history, education background etc.); step 324 to update and enrich the talent profile of the at-risk employee. The talent profile may be created based on the input by the opted-in employee and information received from the HRM system of the candidate supplier. Further, step 324 of workflow 304 may further enrich the talent profile by adding secondary information such as the personal traits and characters based on performance evaluations and peer reviews. Step 324 of workflow 304 may also add information obtained from other third-party sources such as professional social network pages.

[0043] Workflow 304 for the at-risk employees may include steps for interacting with workflow 306 for the job provider, these steps including step 326 for the at-risk employee to receive a job offer from the job provider; step 328 to receive the acceptance (or rejection) or confirmation from the at-risk employee that he/she accepts the job offer; step 330 to notify the job provider of the acceptance of the job offer.

[0044] Workflow 306 may include step 332 for the job provider to create job profiles representing all the jobs that the job provider offers; step 334 to receive groups of potential matching candidates for the job openings (e.g., for each job opening, a group of potential matching talent profiles are provided); when the job provider decides to make a job to an at-risk employee, step 336 to notify the job offer to the at-risk employee’s workflow (step 326); step 338 to update the hire status of the application (e.g., made offer, accepted offer, declined offer etc.); step 340 to track acceptance and hire in the HRM system of the job provider.

[0045] FIG. 5 illustrates a process 400 for an at-risk employee to use the job marketplace according to an implementation of the disclosure. As shown in FIG. 5, at 402, the at-risk employee may visit the landing page of the job marketplace introduced by his or her current employer. The current employer may have placed the employee at risk for a furlough or layoff program. At 404, the at-risk employee may create a candidate account with the job marketplace by selecting a user name and a password. At 406, the at-risk employee may answer enrollment questions (personal information, work history, education background etc.) to create a talent profile for the at-risk employee. At 408, the job marketplace may further update and enrich the talent profile by supplementing the talent profile with secondary information and information collected from other sources. The job marketplace may execute the machine learning modules to determine jobs that match the talent profile of the at-risk employee, and provide a list of jobs to the at-risk employee to review. At 410, the at-risk employee may review the recommended jobs and decide whether to apply for these jobs.

[0046] At 412, the at-risk employee may apply for one or more recommended jobs by answering some job-screening questions. At 414, the at-risk employee may monitor the application status such as communications with the job provider about interviews and offers. At 416, if the at-risk employee received a job offer from the job provider, the at- risk employee may accept the job offer by notifying the job provider (e.g., by sending an acceptance message to the HRM system of the job provider).

[0047] The job provider may correspondingly perform steps for hiring the at-risk employee. At 418, the job provider may create job profiles for job openings. At 420, the job provider may receive talent profiles or enriched talent profiles of candidates for a job. At 424, the job provider may schedule pre-screening interviews with the candidate. If the job provider decides to hire the candidate, at 422, the job provider may notify the candidate with an offer. All of the interactions between the job provider and the candidates may be carried out using interfaces of the job marketplace.

[0048] FIG. 6 illustrates a flowchart of a method 500 for implementing a job marketplace according to an implementation of the disclosure. Method 500 may be performed by processing devices that may comprise hardware (e.g., circuitry, dedicated logic), computer readable instructions (e.g., run on a general purpose computer system or a dedicated machine), or a combination of both. Method 500 and each of its individual functions, routines, subroutines, or operations may be performed by one or more processing devices of the computer device executing the method. In certain implementations, method 500 may be performed by a single processing thread. Alternatively, method 500 may be performed by two or more processing threads, each thread executing one or more individual functions, routines, subroutines, or operations of the method.

[0049] For simplicity of explanation, the methods of this disclosure are depicted and described as a series of acts. However, acts in accordance with this disclosure can occur in various orders and/or concurrently, and with other acts not presented and described herein. Furthermore, not all illustrated acts may be needed to implement the methods in accordance with the disclosed subject matter. In addition, those skilled in the art will understand and appreciate that the methods could alternatively be represented as a series of interrelated states via a state diagram or events. Additionally, it should be appreciated that the methods disclosed in this specification are capable of being stored on an article of manufacture to facilitate transporting and transferring such methods to computing devices. The term “article of manufacture,” as used herein, is intended to encompass a computer program accessible from any computer-readable device or storage media. In one implementation, method 500 may be performed by processing devices 102 implementing job marketplace 108 as shown in FIG. 1.

[0050] As shown in FIG. 6, processing devices 102 may, at 502, identify a talent profile associated with an employee, the talent profile comprising at least one of an employment role held by the employee or a job skill of the employee.

[0051] At 504, processing devices 102 may combine the first set of talent profiles to generate a talent profile collector.

[0052] At 506, processing devices 102 may execute a first neural network module to generate a first- level match score between the talent profile collector and a job profile collector provided by a second information system of a second entity.

[0053] At 508, processing devices 102 may determine whether to provide the first set of talent profiles to the second entity based on the first-level match score. [0054] At 510, responsive to determining to provide the first set of talent profiles to the second entity based on the first-level match score, processing devices 102 may execute a second neural network module to generate a respective second-level match score between each of the first set of talent profiles and at least one job profile in the job profile collector.

[0055] At 512, processing devices 102 may present, in a first user interface, the second- level match scores in a ranked order.

[0056] In some implementations, the candidate supplier may also use the job marketplace 108 to further provide analytics and estimated impacts by the outplacement program. For example, the candidate supplier may use the job marketplace 108 to determine the human capital loss and its impact for a certain furlough program, and make a decision of whether the furlough program is beneficial or detrimental to the business operation.

[0057] FIG. 7 depicts a block diagram of a computer system operating in accordance with one or more aspects of the present disclosure. In various illustrative examples, computer system 600 may correspond to the processing devices 102 of FIG. 1.

[0058] In certain implementations, computer system 600 may be connected (e.g., via a network, such as a Local Area Network (LAN), an intranet, an extranet, or the Internet) to other computer systems. Computer system 600 may operate in the capacity of a server or a client computer in a client-server environment, or as a peer computer in a peer-to-peer or distributed network environment. Computer system 600 may be provided by a personal computer (PC), a tablet PC, a set-top box (STB), a Personal Digital Assistant (PDA), a cellular telephone, a web appliance, a server, a network router, switch or bridge, or any device capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that device. Further, the term "computer" shall include any collection of computers that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methods described herein.

[0059] In a further aspect, the computer system 600 may include a processing device 602, a volatile memory 604 (e.g., random access memory (RAM)), a non-volatile memory 606 (e.g., read-only memory (ROM) or electrically-erasable programmable ROM (EEPROM)), and a data storage device 616, which may communicate with each other via a bus 608.

[0060] Processing device 602 may be provided by one or more processors such as a general purpose processor (such as, for example, a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a microprocessor implementing other types of instruction sets, or a microprocessor implementing a combination of types of instruction sets) or a specialized processor (such as, for example, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), or a network processor).

[0061] Computer system 600 may further include a network interface device 622. Computer system 600 also may include a video display unit 610 (e.g., an LCD), an alphanumeric input device 612 (e.g., a keyboard), a cursor control device 614 (e.g., a mouse), and a signal generation device 620.

[0062] Data storage device 616 may include a non-transitory computer-readable storage medium 624 on which may store instructions 626 encoding any one or more of the methods or functions described herein, including instructions of the job marketplace of FIG. 1 for implementing method 500.

[0063] Instructions 626 may also reside, completely or partially, within volatile memory 604 and/or within processing device 602 during execution thereof by computer system 600, hence, volatile memory 604 and processing device 602 may also constitute machine-readable storage media.

[0064] While computer-readable storage medium 624 is shown in the illustrative examples as a single medium, the term "computer-readable storage medium" shall include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) that store the one or more sets of executable instructions. The term "computer-readable storage medium" shall also include any tangible medium that is capable of storing or encoding a set of instructions for execution by a computer that cause the computer to perform any one or more of the methods described herein. The term "computer-readable storage medium" shall include, but not be limited to, solid-state memories, optical media, and magnetic media. [0065] The methods, components, and features described herein may be implemented by discrete hardware components or may be integrated in the functionality of other hardware components such as ASICS, FPGAs, DSPs or similar devices. In addition, the methods, components, and features may be implemented by firmware modules or functional circuitry within hardware devices. Further, the methods, components, and features may be implemented in any combination of hardware devices and computer program components, or in computer programs.

[0066] Unless specifically stated otherwise, terms such as “receiving,” “associating,” “determining,” “updating” or the like, refer to actions and processes performed or implemented by computer systems that manipulates and transforms data represented as physical (electronic) quantities within the computer system registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices. Also, the terms "first," "second," "third," "fourth," etc. as used herein are meant as labels to distinguish among different elements and may not have an ordinal meaning according to their numerical designation.

[0067] Examples described herein also relate to an apparatus for performing the methods described herein. This apparatus may be specially constructed for performing the methods described herein, or it may comprise a general purpose computer system selectively programmed by a computer program stored in the computer system. Such a computer program may be stored in a computer-readable tangible storage medium.

[0068] The methods and illustrative examples described herein are not inherently related to any particular computer or other apparatus. Various general purpose systems may be used in accordance with the teachings described herein, or it may prove convenient to construct more specialized apparatus to perform method 300 and/or each of its individual functions, routines, subroutines, or operations. Examples of the structure for a variety of these systems are set forth in the description above.

[0069] The above description is intended to be illustrative, and not restrictive. Although the present disclosure has been described with references to specific illustrative examples and implementations, it will be recognized that the present disclosure is not limited to the examples and implementations described. The scope of the disclosure should be determined with reference to the following claims, along with the full scope of equivalents to which the claims are entitled.

Claims

CLAIMS What is claimed is:

1. A computing system implementing a multi-level job marketplace, the computing system comprising: a memory device; and one or more processing devices, communicatively connected to the memory device, to: identify, based on one or more features, a first set of talent profiles from a pool of talent profiles supplied by a first information system of a first entity; combine the first set of talent profiles to generate a talent profile collector; execute a first neural network module to generate a first-level match score between the talent profile collector and a job profile collector provided by a second information system of a second entity; and determine whether to provide the first set of talent profiles to the second entity based on the first-level match score.

2. The system of claim 1, wherein the one or more processors are further to: responsive to determining to provide the first set of talent profiles to the second entity based on the first- level match score, execute a second neural network module to generate a respective second-level match score between each of the first set of talent profiles and the at least one job profile in the job profile collector; and present, in a first user interface, the second-level match scores in a ranked order.

3. The system of any of claim 1 or 2, wherein the one or more features comprise at least one of a job title, a team identifier, a project identifier, a job skill, or an education achievement, wherein the talent profile collector comprises talent profiles of the first entity, and wherein the job profile collector comprises job openings associated with at least one of a job title, a team identifier, or a project identifier of the second entity.

4. The system of any of claim 1 or 2, wherein each of the pool of talent profiles is associated with a corresponding employee of the first entity, and wherein each of the pool of talent profiles comprises at least one category entry, the at least one category entry comprising one or more data items.

5. The system of claim 4, wherein to combine the first set of talent profiles to generate a talent profile collector, the one or more processing devices are to: combine data items in each of the set of talent profiles according to category entries in the talent profile collector; or combine data items in each of the set of talent profiles into the talent profile collector as one container data object.

6. The system of any of claim 1 or 2, wherein the one or more processing devices are further to: receive the neural network module trained by adjusting at least one parameter associated with the first neural network module using a training talent profile collector constructed from a training set of talent profiles, and a training job profile collector constructed from a training set of job profiles.

7. The system of any of claim 1 or , wherein to determine whether to provide the first set of talent profiles to the second entity based on the first-level match score, the one or more processing devices are to: compare the first-level match score with a threshold value; and responsive to determining that the first- level match score is equal to or greater than the threshold value, filter the first set of talent profiles and transmit the set of the filtered talent profiles to the second information system of the second entity.

8. The system of any of claim 1 or 2, wherein the one or more processing devices are further to: execute the first neural network module to generate a second first-level match score between the talent profile collector and a second job profile collector provided by a third information system of a third entity; and determine whether to provide the first set of talent profiles to the second entity or the third entity based on a comparison between the first- level match score and the second first-level match score.

9. The system of claim 8, wherein responsive to determining that the second first- level match score is larger than the first- level match score, the one or more processing devices are to provide the set of talent profiles to the third information system of the third entity.

10. The system of claim 2, wherein the one or more processing devices are to: receive the second neural network module trained by adjusting at least one parameter associated with the second neural network module using a training talent profile and a training job profile.

11. A method for implementing a multi-level job marketplace, the method comprising: identifying, by one or more processing devices based on one or more features, a first set of talent profiles from a pool of talent profiles supplied by a first information system of a first entity; combining, by the one or more processing device, the first set of talent profiles to generate a talent profile collector; executing, by the one or more processing devices, a first neural network module to generate a first- level match score between the talent profile collector and a job profile collector provided by a second information system of a second entity; and determining, by the one or more processing devices, whether to provide the first set of talent profiles to the second entity based on the first-level match score.

12. The method of claim 11, further comprising: responsive to determining to provide the first set of talent profiles to the second entity based on the first- level match score, executing a second neural network module to generate a respective second-level match score between each of the first set of talent profiles and the at least one job profile in the job profile collector; and presenting, in a first user interface, the second-level match scores in a ranked order.

13. The method of any of claim 11 or 12, wherein the one or more features comprise at least one of a job title, a team identifier, a project identifier, a job skill, or an education achievement, wherein the talent profile collector comprises talent profiles of the first entity, and wherein the job profile collector comprises job openings associated with at least one of a job title, a team identifier, or a project identifier of the second entity.

14. The method of any of claim 11 or 12, wherein each of the pool of talent profiles is associated with a corresponding employee of the first entity, and wherein each of the pool of talent profiles comprises at least one category entry, the at least one category entry comprising one or more data items.

15. The method of claim 14, wherein combining, by the one or more processing device, the first set of talent profiles to generate a talent profile collector further comprises: combining data items in each of the set of talent profiles according to category entries in the talent profile collector; or combining data items in each of the set of talent profiles into the talent profile collector as one container data object.

16. The method of any of claim 11 or 12, further comprising: receiving the neural network module trained by adjusting at least one parameter associated with the first neural network module using a training talent profile collector constructed from a training set of talent profiles, and a training job profile collector constructed from a training set of job profiles.

17. The method of any of claim 11 or 12, wherein determining, by the one or more processing devices, whether to provide the first set of talent profiles to the second entity based on the first- level match score comprises: comparing the first-level match score with a threshold value; and responsive to determining that the first- level match score is equal to or greater than the threshold value, filtering the first set of talent profiles and transmitting the set of the filtered talent profiles to the second information system of the second entity.

18. The method of any of claim 11 or 12, further comprising: executing the first neural network module to generate a second first-level match score between the talent profile collector and a second job profile collector provided by a third information system of a third entity; and determining whether to provide the first set of talent profiles to the second entity or the third entity based on a comparison between the first- level match score and the second first-level match score.

19. The method of claim 18, further comprising: responsive to determining that the second first-level match score is larger than the first-level match score, providing the set of talent profiles to the third information system of the third entity.

20. A machine-readable non-transitory storage media encoded with instructions that, when executed by one or more processing devices, cause the one or more processing devices to implement a system a multi-level job marketplace, to: identify, based on one or more features, a first set of talent profiles from a pool of talent profiles supplied by a first information system of a first entity; combine the first set of talent profiles to generate a talent profile collector; execute a first neural network module to generate a first-level match score between the talent profile collector and a job profile collector provided by a second information system of a second entity; and determine whether to provide the first set of talent profiles to the second entity based on the first-level match score.