Browse Prior Art Database

Dynamically building a skills profile based on a user's contributions to company-specific applications using language processing.

IP.com Disclosure Number: IPCOM000246913D
Publication Date: 2016-Jul-14
Document File: 3 page(s) / 64K

Publishing Venue

The IP.com Prior Art Database

Abstract

Dynamically building a skills profile based on a user's contributions to company-specific applications. Overall scores in a skill area are calculated by aggregating weighted scores, based on response, importance and quantity using language processing.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 49% of the total text.

Page 01 of 3

Dynamically building a skills profile based on a user's contributions to company-specific applications using language processing.

Large companies often require their employees to maintain a skills profile. These are used to connect people and facilitate the sharing of knowledge.The common implementation requires a user to input data on their experience and skill set. The issue is that the information is only as accurate as the individual says it is, and it requires constant maintenance to keep current. This can lead to confusion and time delays when searching for specific domain knowledge. This idea solves the problem of user-defined, out-of-date, and non-validated skills profiles.

    Every employee will have a skills profile developed through semantic analysis of work-related contributions. This will be done using plugins to pull data periodically from relevant applications about the topics that an individual frequently contributes to, and will give a score based on response, importance and quantity. This idea improves on the solutions listed above in that the skills are not manually generated, and are continually updated, giving a more reliable view of an individual's expertise.

    Each user has a skills profile which is made up of various experience scores depending on their previous or current work. These scores are calculated using contextual analysis of their contributions within, for example, work items, correspondence, wiki pages, communities, and other company specific tools and applications. This is a series of plugins which collect data from a variety of sources. It is a continuous process. On initial discovery, the plugins pull all past data and pass it to the Semantic Analysis Engine (SAE) which creates a profile. Subsequently, the profile is honed without any interaction from the user, and skills are added or adjusted as new contributions are made. Consequently, the skills profile is constantly developing and improving.

    These plugins communicate with the SAE via an API. The SAE uses learned key words to perform linguistic analysis. This helps to decide how knowledgeable each person is, taking into account how regularly someone contributes to a topic and how recently they have contributed. As an extra level, sentiment analysis provides a method for assessing the quality of the data by assessing responses.

    When dealing with a large company's data, confidentiality is an important issue. This would be tackled by having a list of company-specific excluded key

words which would not appear in any profile.

    This idea removes the need for colleagues to rate each other's work and uses various levels of analysis to create a realistic picture of the skills of an individual. From here, the skills profile can be used to recommend experts based on their scores. This solution is intelligent enough to need no user interaction after installation.

    The novel part of this idea is the way in which contributions are weighted and added to an over...