Faysal AzizFA-0716Data & Analytics
KNIME Essentials for Spreadsheet Users
An abridged beginner introduction to visual data workflows
Introduction
Why this course
Use KNIME Analytics Platform alongside spreadsheet work to prepare, combine and inspect small datasets through visual workflows. This abridged course assumes no prior programming knowledge and concentrates on practical data-handling foundations.
KNIME is a workflow platform rather than an Excel add-in. Prediction examples are introductory demonstrations with evaluation, not promises of accurate forecasts or accelerated work in every case.
Learning outcomes
Learning outcomes
- Navigate the workspace and connect, configure and execute workflow nodes.
- Read spreadsheet data and inspect column types and missing values.
- Filter, concatenate, join and aggregate selected datasets.
- Apply basic discretisation, normalisation, pivoting and data cleaning.
- Create a simple visualisation and explain the results.
- Recognise the inputs, evaluation needs and limits of introductory k-nearest-neighbour, XGBoost and time-series examples.
Prerequisites
Prerequisites
- Familiarity with using a personal computer.
- Comfortable using the internet.
- Some basic understanding of FIle systems.
Access to an approved KNIME installation, compatible example workflows and any extensions required for the supplied demonstrations.
Training outline
2 modules
·
01Day 1 — Visual workflows and spreadsheet data5 topics
- Explore KNIME Analytics Platform, the local workspace, node repository and workflow editor.
- Connect, configure, execute and reset nodes; inspect input/output tables and errors.
- Read small Excel datasets, examine data types and combine multiple sources.
- Filter rows or columns, concatenate compatible tables, join on keys and aggregate values.
- Practical work: build and document a short data-preparation workflow; check row counts and join assumptions.
02Day 2 — Preparation, views and prediction awareness6 topics
- Clean selected values and handle missing data; introduce discretisation, normalisation and pivoting.
- Use a small loop example where repetition is useful and inspect its behaviour.
- Create a chart and check whether it supports the intended interpretation.
- Explore supplied k-nearest-neighbour and XGBoost examples using appropriate extensions and separate evaluation data.
- Introduce a bounded time-series example, chronological evaluation and the limits of forecasting.
- Review the prepared workflow and identify where further data or modelling knowledge is required.
A programme built around your team.
Share your training goals and requirements.