If there is not enough information, It may be required to acquire some data from other sources through the usage of web crawlers, and the storage of this data into databases or spreadsheets. This process is part of the data mining process, and some Python tools used with this purpose are SELENIUM, SCRAPY, REQUESTS, BS4 and some languages like XPATH.
Through this process, the data is inspected, cleaned, transformed and modelated to extract useful information that helps the decision making. "Data is... analyzed to answer questions, test hypotheses, or disprove theories". Statistical knowledge, Python tools to explore and transform data like pandas, and tools like SciPy or Scikit-learn help to conclude and communicate the findings inside the data.
This process aims to show the findings obtained through the data analysis, using a practical method relying on the target audience.
Some tools used for technical topics are matplotlib and seaborn; but for a better dataviz can be used Plotly, Tableau, DataStudio or PowerBI.
Hi! I'm Herman, an electronic engineer and master of business administration.
My experience managing projects, taught me the importance of transforming data into information, to improve the performance in the company processes, operations and projects.
This inspired me to fully dedicate myself to data analysis, giving me the opportunity to help companies in their success and to guide them into a data driven philosophy.
This project looks for a classification model to identify if a user would stop paying a loan (defaulter user is identified with the label "1"). Metrics like "recall" or "F1" will show how trustable the model is, furthermore the Logistic Regression model will allow us to visualize which variables are more important to identify a defaulter user.
This project shows step by step the process from EDA to modeling a model to forecast the sales of a product in a period of time.
It contains vectorization of data, clustering, and forecasting models like Decission trees, Random forest and XGBoost.
This project shows how to get some data from the internet to transform it into useful information that can help in the decision making processes.
The final result is a visualization in Tableau letting the user interact with the findings of the data.
For this project it was used Python, selenium, descriptive statistics and clustering algorithms.
This is a brief notebook which shows the implementation of probability basics to make an analysis about a problem, using some basic programming tools in Python.
This project aims to scrape with Selenium a maximum of 500 vacancies from Linkedin,
and then create a spreadsheet with the respective description of the job vacancies.
Furthermore, this spreadsheet file is processed by another Python script that will
create a dataviz to show which abilities are most demanded based on the data of the file.
This project tries to estimate if a target variable is
affected by some other predictor variables, to get a conclusion.
This project is under construction.