System and method for evaluating human resources in a software development environment
The present disclosure discloses system and method for evaluating a human resource in a software development environment. At first, historical performance data and profile data associated with a plurality of human resources involved in a software project is received. From such data (historical performance data and profile data), a plurality of attributes is extracted. Further, Bayesian classification technique is implemented on the plurality of attributes in order to classify the plurality of attributes, of each human resource, into a plurality of classes. The plurality of attributes is classified in such a manner that at least one attribute corresponding to at least one human resource and at least one other human resource is classified into a class and another class respectively. Further, based on the classification of each attribute associated with the human resource, a grade is assigned to the human resource.
1. A method for automatically evaluating a human resource in a software development environment, wherein the method comprising:
receiving, by a processor, historical performance data and profile data, associated with a plurality of human resources involved in a software project, as raw data,
wherein the raw data is received from a plurality of repositories comprising user profile system Subversion (SVN) repository, defect repositories, and project tracking system;
retrieving, by the processor, a plurality of attributes, from the historical performance data and the profile data, corresponding to each human resource;
classifying, by the processor, the plurality of attributes, of each human resource, into a plurality of classes by implementing a Bayesian classification technique on the plurality of attributes,
wherein the Bayesian classification technique comprises a discriminative probabilistic classifier model and a generative probabilistic model, and
wherein the plurality of attributes are classified, by using at least one of the discriminative probabilistic classifier model and the generative probabilistic model, such that at least one attribute corresponding to at least one human resource and at least one other human resource is classified into a class and another class respectively; and
assigning, by the processor, a grade to a human resource based on the classification of each attribute associated with the human resource, thereby automatically evaluating the human resource in the software development environment.
2. The method of claim 1 , wherein the plurality of attributes comprises total years of experience, years of experience relevant for the software project, code completion duration, number of defects fixed, number defects reported corresponding to the developed code, complexity, functional complexity, thousands of lines of code (KLOC), location of the human resource, self-assessment rating given by the human resource, and a supervisor-rating given by a project-supervisor of the software project.
3. The method of claim 1 , further comprising allocating a task, associated with the software project, to a human resource, of the plurality of human resources, based on the grade assigned to the human resource.
4. The method of claim 1 , wherein the grade assigned is at least one of an excellent, above average, average, and below average.
5. The method of claim 1 further comprises
collecting, by the processor, the raw data from the plurality of repositories; and
preparing, by the processor, the raw data for extraction of the plurality of the attributes.
6. A system for evaluating a human resource in a software development environment, the system comprises:
a processor;
a memory coupled to the processor, wherein the processor executes a set of instructions stored in the memory to:
receive historical performance data and profile data, associated with a plurality of human resources involved in a software project, as raw data,
wherein the raw data is received from a plurality of repositories comprising user profile system Subversion (SVN) repository, defect repositories, and project tracking system;
retrieve a plurality of attributes, from the historical performance data and the profile data, corresponding to each human resource;
classify the plurality of attributes, of each human resource, into a plurality of classes by implementing a Bayesian classification technique on the plurality of attributes,
wherein the Bayesian classification technique comprises a discriminative probabilistic classifier model and a generative probabilistic model, and wherein the plurality of attributes are classified, by using at least one of the discriminative probabilistic classifier model and
the generative probabilistic model, such that at least one attribute corresponding to at least one human resource and at least one other human resource is classified into a class and another class respectively; and
assign a grade to a human resource based on the classification of each attribute associated with the human resource, thereby automatically evaluating the human resource in the software development environment.
7. The system of claim 6 , wherein the plurality of attributes comprises total years of experience, years of experience relevant for the software project, code completion duration, number of defects fixed, number defects reported corresponding to the developed code, complexity, functional complexity, thousands of lines of code (KLOC), location of the human resource, self-assessment rating given by the human resource, and a supervisor-rating given by a project-supervisor of the software project.
8. The system of claim 6 , further comprise to allocate a task, associated with the software project, to a human resource, of the plurality of human resources, based on the grade assigned to the human resource.
9. The system of claim 6 , wherein the grade assigned is at least one of an excellent, above average, average, and below average.
10. The system of claim 6 is further configured to
collect the raw data from the plurality of repositories; and
prepare the raw data for extraction of the plurality of the attributes.
11. A non-transitory computer readable medium embodying a program executable in a computing device for evaluating a human resource in a software development environment, the program comprising:
a program code for receiving historical performance data and profile data, associated with a plurality of human resources involved in a software project, as raw data, wherein the raw data is received from a plurality of repositories comprising user profile system Subversion (SVN) repository, defect repositories, and project tracking system;
a program code for retrieving a plurality of attributes, from the historical performance data and the profile data, corresponding to each human resource;
a program code for classifying the plurality of attributes, of each human resource, into a plurality of classes by implementing a Bayesian classification technique on the plurality of attributes,
wherein the Bayesian classification technique comprises a discriminative probabilistic classifier model and a generative probabilistic model, and
wherein the plurality of attributes are classified, by using at least one of the discriminative probabilistic classifier model and the generative probabilistic model, such that at least one attribute corresponding to at least one human resource and at least one other human resource is classified into a class and another class respectively; and
a program code for assigning a grade to a human resource based on the classification of each attribute associated with the human resource, thereby automatically evaluating the human resource in the software development environment.