Abstract

In many countries NIS are facing decreasing response rates and increasing survey costs.
Alternative sampling and recruiting approaches are usually needed, including non-probability and online sampling Because of the massive popularity of social networks, data about the users and their communication offers unprecedented opportunities to examine how human society functions at scale
The aim of this paper is to show how linkedin is as well as the most comprehensive and up-to-date professional database, a real mine of information on education and on the the labour market and at the same time and alternative source to labour force surveys conducted by statistical institutes.
Based on.web scraping conducted without violating LinkedIn's terms of service and data privacy regulations, methodology includes Data Preprocessing, Feature Extraction then a basic workflow using TF-IDF for feature extraction and an SVM classifier and Model Definition for Classification that Use supervised learning algorithms such as Support Vector Machines (SVM), Naive Bayes, Decision Trees, Random Forests, or neural networks like Multi-layer Perceptron (MLP). Then, Model Evaluation, Optimization and Validation
The data collected in this way provides a picture of labour demand and supply, i.e. the demographic characteristics of LinkedIn users, the profile of the companies (sector, size, etc.) offering jobs
The results will then be compared with those of a traditional survey to detect differences, strengths and weaknesses.