Data analyzing and machine learning Hands-on with KNIME
- 3.5 hours on-demand video
- 18 downloadable resources
- Full lifetime access
- Access on mobile and TV
- Certificate of Completion
Get your team access to Udemy's top 3,000+ courses anytime, anywhere.Try Udemy for Business
- create machine learning models in Knime Analytics Platform from A to Z – classification and regression
create machine learning models - Regression (simple linear, multilinear, polynomial, decision tree, random forest, gradient booster)
create machine learning models - Classification (decision tree, random forest, naive bayes, KNN, gradient booster)
- prepare the data for the machine learning predictive model by using basic manipulating KNIME nodes
- Evaluate the performance of the machine learning predictions (confusion matrix, accuracy ratio, scatter plot)
- work with several different file’s data sources at one place
- work with the workflow files and Knime nodes
- acquire data into the Knime workflow
- manipulate the data by using basic Knime nodes
- visualize the data by using plots and statistics Knime nodes
- understand the basic theory of the machine learning
- install and understand the Knime Analytics Platform environment
- find help and advice when working with Knime
- access to computer or laptop with Windows (32bit or 64 bit), Linux (64bit) or Mac (64bit) and with permission to download softwares (if not, ask your administrator to download it for you – it is common at company´s computers)
- no prior knowledge required
- basic data analyzing experience in different programs, like MS Excel or SQL or Python etc. is added advantage
The goal of this course is to gain knowledge how to use open source Knime Analytics Platform for data analysis and machine learning predictive models on real data sets.
We will create machine learning models within the standard machine learning process way, which consists from:
- acquiring data by reading nodes into the KNIME software (the data frames are available in this course for download)
- pre-processing and transforming data to get well prepared data frame for the prediction
- visualizing data with KNIME visual nodes (we will create basic plots and charts to have clear picture about our data)
- creating machine learning predictive models and evaluating them:
1. Decision Tree Classification
2. Simple linear Regression
from 11.2.2019 newly added:
3. Decision Tree Regression
4. Random Forest Regression
5. Random Forest Classification
6. Polynomial Regression (+ info about multi linear Regression - for Knime same nodes)
7. Naive Bayes
8. K nearest neighbors
9. Grandient booster Regression
10. Grandient booster Classification
I will also explain the Knime Analytics Platform environment, guide you through the installation and show where to find help.
- anyone searching user-friendly, easily understandable and highly useful tool for data analyzing and machine learning tasks without necessity to have programming skills
- people working with several data sources of different file types
- people working with data - both small and big data
- anyone excited in learning new things in the data science field
- people willing to learn and use new modern tools for data analyzing and machine learning
We will guide you through the installation of the Knime analytics platform
We will create new workflows we will us for our machine learning predictive models.
After watching this video, please download three uploaded excel files.
We will create the decision tree classification model based on already prepared data we have done in last chapters. Also, we create the model to be able predict data when new data available.
We will create the linear regression model based on already prepared data we have done in last chapters. Your task will be to finalize the model to be able predict data when new data available.
In the next lecture you can check your result.
In this lecture we will create second machine learning technique from the ensemble learning group, which is the gradient boost model (and random forest we have already done).
Also, we will create for each regression model prediction the scatter plot to evaluate the single models in the chart form to easily read where the prediction was right and where wrong.