Skip to content

predict which workers are most likely to leave the company using logistic regression, random forest and neural networks with Tensor Flow

License

Notifications You must be signed in to change notification settings

bertuci/human_resources_department

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Human Resources Department

predict which workers are most likely to leave the company using logistic regression, random forest and neural networks with Tensor Flow

Screenshot_47

CONTEXT OF THE PROJECT

. You work as a data scientist for a company

. The HR department has collected data from employees and would like you to predict which ones are most likely to leave your job

. Some data examples:

  • Involvement with work
  • Education
  • Job satisfaction
  • Performance metrics
  • Relationship
  • Balance between personal and professional activities

Data source: https://www.kaggle.com/pavansubhasht/ibm-hr-analytics-attrition-dataset

INFORMATION RELATING TO THE PROBLEM OF EMPLOYEES LEAVE THE COMPANY.

• Hiring and retaining employees are extremely complex tasks that require capital, time and skills

• "Small business owners spend around 40% of their working hours on tasks that do not generate revenue, such as hiring"

• “Companies spend 15% to 20% of employees' salaries to recruit a new candidate

• "An average company loses between 1% and 2.5% of its total revenue in the time it takes to train a new employee"

• Hiring a new employee costs an average of $ 7645 (in a company with approximately 500 employees)

• It takes about 52 days for an employee to actually occupy his new position

Source: https://toggl.com/blog/cost-of-hiring-an-employee

TECHNIQUES USED FOR THIS PROBLEM

• Logistic regression • Random Forest • Neural Networks (using the Tensor Flow)

AGORA EM BR

CONTEXTO DO PROJETO

. Você trabalha como cientista de dados para uma empresa

. O departamento de RH coletou dados dos funcionários e gostaria que você fizesse a previsão de quais são mais propensos para sair do emprego

. Alguns exemplos de dados:

  • Envolvimento com trabalho
  • Escolaridade
  • Satisfação com o trabalho
  • Métricas de desempenho
  • Relacionamento
  • Equilíbrio entre atividades pessoais e profissionais

Fonte dos dados: https://www.kaggle.com/pavansubhasht/ibm-hr-analytics-attrition-dataset

INFORMAÇÕES RELACIONADAS AO PROBLEMA DOS COLABORADORES SAIREM DA EMPRESA.

• Contratar e reter funcionários são tarefas extremamente complexas que exigem capital, tempo e habilidades

• “Pequenos empresários gastam em torno de 40% das horas de trabalho em tarefas que não geram receitas, comoa contratação”

• “Empresas gastam de 15% a 20% do saláriodos funcionários para recrutarum novo candidato

• “Uma empresa média perde entre 1% e 2.5% de sua receita total no tempo que leva para treinar um novo funcionário”

• A contratação de um novo funcionário custa em média $7645 (em uma empresa com aproximadamente 500 funcionários)

• Demora mais ou menos 52 dias para um funcionário ocupar de fato sua nova posição

Fonte: https://toggl.com/blog/cost-of-hiring-an-employee

TÉCNICAS UTILIZADAS PARA ESSE PROBLEMA

• Regressão Logística • Random Forest •Redes Neurais(utlizando o tensor flow)

About

predict which workers are most likely to leave the company using logistic regression, random forest and neural networks with Tensor Flow

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published