Expert system for recruitment. Development of an expert system for recruitment based on a neural network
It's obvious that software system will not be able to independently understand that this or that document corresponds to the information needs of the user. But what if a person, and in this case an expert, tells him...
Methods for automatically generating search heuristics
2. Automatic construction of a set of heuristics 3...
Methods for automatically generating search heuristics
The expert system is trained independently for each pair (rubric, type of resource). Let's introduce the following definitions used in the process of training an expert system: Relevant text is text...
Overview of expert systems
Information complexity, structural change and an increase in its volume create new the necessary conditions to processing and, as a result, the speed of solving some problems that cannot be solved without special knowledge...
Internal database predicate description section
To conduct an examination computer program must be able to solve problems by means of inference and at the same time obtain sufficiently reliable results. The program must have access to a system of facts called the knowledge base...
Development automated system personnel management ACS "HR Department"
Figure 2.1 shows the generalized structure of an expert system. Figure 2.1 - Typical structure of an expert system It should be noted that real ES may have a more complex structure, however, the blocks shown in the figure ...
Development of an expert system shell
Development of an expert system for recruitment based on a neural network
An expert system (ES) is a system that generates decisions based on a knowledge base (KB) (in particular, on production ones). Production system (PS) or Products \u003d (RDB, MV, BZ) The general structure of the expert system (ES) is shown in the figure ...
Creation of an expert system for the subject area "Construction of a bath on suburban area"
Combining information about the properties of objects and relations between them, we obtain a set of facts that should be included in the working database (RDB) of the intelligent system: RDB = ( Bath_1.Constructed [ Fact type: Boolean; Initial value = false ]...
Expert system "Sports equipment selection"
The program is an interactive shell that can ask the user questions. When the program asks a question, the user can answer "Yes" or "No"...
Expert system for automating decision-making processes when choosing a motorcycle
The developed expert system will consist of several modules, development languages html, php. On the html language written markup of the system pages, controls, questions module...
Expert systems
The generalized structure of the expert system is shown in the figure. It should be noted that real EIS may have a more complex structure, but the blocks shown in the figure are certainly present in any real expert system...
To streamline the recruitment process, some US companies are using computerized expert systems to pre-interview applicants for jobs. According to the majority of users, such systems help to increase the completeness and reliability of the data received, help to evaluate candidates more objectively, and positively affect the performance of enterprises (productivity, staff turnover, the level of violations, theft, etc.). The expert system concentrates the most effective methods of work accumulated by the company's personnel department specialists, allowing them to overcome such human shortcomings as absent-mindedness, excessive emotionality, slowness, and fear of posing a “sharp question” to the interlocutor. According to the observations of psychologists, respondents, when “talking” with a computer, experience less excitement and give more frank answers than when communicating with a live interviewer.
Interviewing and testing programs for people applying for a job should be, in the terminology of computer scientists, "friendly", i.e. provide a natural way for the user to interact, protect against errors, developed tools for prompts and interactive documentation. In particular, they may include questions with mutually exclusive answers. The nature of the questions, as well as their sequence during the interview, are individualized depending on the content of the answers. The computer controls the content of the respondent's answers, draws his attention to emerging contradictions, fixes the speed of response to individual questions.
Such a program contains an average of 75-125 questions formulated taking into account the job requirements and the specifics of the company's activities and providing for mutually exclusive answers. The duration of testing is about 20 minutes; it would take an HR officer about two hours to obtain the same amount of information in a face-to-face interview. The obtained results are then examined by various methods of psychometric and statistical analysis and compared with the data of other candidates.
At the end of the interview, the computer gives the HR employee a tabulagram summarizing the results of the interview in sections: summary data (include a breakdown of answers by given key topics: previous work activity, life experience); a systematic list of contradictions in the answers; speed of response (questions are highlighted, answering which the respondent spent time exceeding the average for the test); a list of “problem” answers (requiring additional analysis and specification during a subsequent interview with an employee of the personnel service); a sample list of questions that a recruiter should ask during an interview.
In addition, if provided by the program, the results of testing certain psychological or business qualities of the candidate are given according to his answers.
Summary
1. All work on the selection, training, placement and rational use of personnel in production requires an assessment of managerial personnel. The reserve of managerial personnel is understood as a specially formed group of employees, whose qualities meet the requirements that apply to managers of a certain rank, who have passed certification selection, special managerial training and have achieved positive results in production activities. Methods for selecting candidates for the reserve of managerial personnel are divided into three main groups: predictive, practical and laboratory. The list of positions to which a leader of a certain rank has the right to appoint is called his nomenclature. The placement of personnel in managerial positions is the main goal of the company's personnel policy and a key link in all work with management personnel.
Questions and tasks for repetition
1. Describe the main periods and the procedure for certification of managers and specialists. What does the current assessment of management personnel consist of? name general principles reserve work. What are the new requirements for HR managers? Describe the main sources of the leadership pool.
Introduction
Increasingly, we are talking not about point, individual selection, but about mass recruitment, when you need to select many candidates at the same time in a short time. And to do this not only quickly, but also as efficiently as possible (and even at minimal cost). Imagine how many difficulties you need to overcome in order to realize such a big project. It's about managing a huge amount of information, a lot of resources and channels. Not every company, HR department can handle such a task. And not every system software Recruitment will deal with it. Often, personnel officers simply do not have the necessary tools, experience and knowledge to implement such projects.
An expert system is a system that combines the power of a computer and knowledge, together with the experience of an expert, in such a way that it can give reasonable advice or carry out a reasonable solution to a given problem. An additional property of an expert system is the ability to explain the course of one's reasoning in a form understandable to the user.
To do this, it is necessary to develop a software tool, that is, an "Expert system for the selection of personnel using a neural network".
This term paper is to study the system of work with personnel records for the selection of candidates for a particular position.
Benefits and advantages of automating the selection of employees using the "Expert System":
· Reduction of labor costs associated with the input of information;
· Speeding up the search and attraction of candidates.
· Constant growth of base.
· Simplified accounting, exclusion of data loss.
recruitment expert system
· High-quality, full-fledged study of information about the requirements for the vacancy.
Description of the expert system for recruitment
Description for developing a problem statement
As a rule, expert systems are created to solve practical problems in some highly specialized areas, where the knowledge of specialists plays an important role. Expert systems were the first developments that were able to draw a lot of attention to the results of research in the field of artificial intelligence.
Expert systems have one big difference from other artificial intelligence systems: they are not designed to solve any universal problems, expert systems are designed for high-quality problem solving in an area defined by developers, in rare cases - areas.
The basis for the development of the program is the implementation of course work on the discipline " Intelligent systems and technology" on the example of "Expert system for recruitment using a neural network". The program should be developed to optimize recruitment.
Purpose of development
The expert system is designed to store and process information about the selection of personnel. The software tool "Expert system for the selection of personnel based on a neural network" can significantly simplify the search task using the characteristics of the object of information about the object itself, which has these characteristics. This software tool is designed not only for personnel records, it also has the ability to be applied in other areas of activity.
Reliability Requirements
The reliability of the system as a whole depends on the reliability of the operating system used.
The expert system must find a solution within a reasonable time that would be no worse than the one that can be offered by a specialist in this subject area.
Program Requirements
The main requirement for the "Expert Neural Network Recruitment System" is the function of recruiting with the help of a factor.
Performance Requirement
The program must provide the ability to perform the following functions:
Change the characteristics of the object and the object itself;
· Add/change, delete the characteristics of the object and the object itself;
· Save/load the base for storing information about the object;
· Carry out network training;
· Implementation of the recruitment process.
Requirements for the composition and parameters of computer hardware
The computer must be a computer with at least the following characteristics:
· Intel processor Pentium 1000 MHz;
· volume random access memory- 512 MB;
HDD capacity - 30 GB.
As operating systems can be used:
· Microsoft Windows XP and up.