International Business Machines Corporation

David Glenn George

John Arthur Medicke

Scott N Gerard

Stephen Mitchell

Feng-wei Chen

Sudhanshu Shekhar Singh

Joydeep Mondal

Samuel E Reynolds

Sarthak Ahuja

Abhay Choudhary

Matthew Morehouse

Paul Charles James Dunning

Smitashree Choudhury

Amanda Klabzuba

Wing Yin Leung

Jacek Adam Piskorski

A method, computer system, and a computer program product for semantic matching is provided. The present invention may include pre-processing and normalizing a job title. The present invention may include deconstructing the job title based on at least one semantic element. The present invention may include training a machine learning model. The present invention may include creating a contextual word representation of the job title using the at least one semantic element of the job title. The present invention may include computing a similarity score for each of the at least one semantic element of the job title. The present invention may lastly include applying a weight to the computed similarity score before making a final match assessment.

Semantic matching of job titles with limited contexts

A computer-implemented method, a computer program product, and a computer system for determining skill adjacencies using a machine learning model. A computer calculates first similarity scores between first skill vectors obtained from one or more training datasets and second similarity scores between the first skill vectors and skill category vectors pre-calculated from job requisitions, using both a reference corpus word embedding model and a target corpus word embedding model. The computer generates features extracted from the first similarity scores and the second similarity scores. Based on the features, the computer trains a machine learning model for classifying combinations of skills as adjacent and non-adjacent. The machine learning model is used to determine skill adjacencies between skills extracted from the job requisitions and skills extracted from resumes.

Determining skill adjacencies using a machine learning model

A method, system and computer program product for assessing similarity between two job description documents. Job description documents consist of sentences framed in a particular manner, where the sentences are represented as a set of actions, an object corresponding to each action and a set of attributes corresponding to the object. The two job description documents are parsed to generate a first and a second set of an action-object-attribute triplet representation, where the first set of the action-object-attribute triplet representation is associated with the first job description document and the second set of the action-object-attribute triplet representation is associated with the second job description document. A similarity score between the first and second sets of action-object-attribute triplet representations is then calculated by hierarchically matching the first and second sets of action-object-attribute triplet representations across the job description documents. In this manner, similar job positions/job descriptions may be more accurately identified.

Creating and using triplet representations to assess similarity between job description documents

In some embodiments, a method comprises receiving a tag cloud including tags that hyperlink to web content. The method can also comprise separating the tags into different linguistic categories, assigning a weight to each tag, and grouping the tags into clusters, wherein tags in a cluster are associated with a context. The method can also include determining one or more domains for the tag clusters, wherein a domain is a broadest class that defines one or more of the tags in a linguistic category, determining a hierarchy for the tags based on the weights of the tags, and identifying linguistic relationships between the tags. The method can also comprise determining properties associated with one or more of the tags and one or more of the domains, wherein the tag's properties are determined using linguistic analysis and storing the tags, the hierarchies, the linguistic relationships, and the properties.

Deriving ontology based on linguistics and community tag clouds

A timing control mechanism designed for a high performance network wherein many of the timers are cancelled or reset prior to expiring. This method and apparatus enable the cancels and resets to occur with minimal impact to the system performance. An intermediate work queue is used, where timer events are posted prior to being scheduled into the timer queues by the timer process. Thus, many of the cancels or resets will be implemented simply by changing the status of events in this work queue, minimizing the impact on the timer queues and timers. The timer process processes the work queue before processing any timer events.

Timer state control optimized for frequent cancel and reset operations

Growing community of inventors

Cary, NC, United States of America

David Glenn George