Creating a customer segmentation workflow with knime
0 likes313 views
The document discusses creating a customer segmentation workflow using the Knime analytics platform, including introducing Knime, explaining what customer segmentation is and its benefits, demonstrating how to perform customer segmentation using k-means clustering in Knime, and providing a demo of customer segmentation for Domino's pizza. The workflow would involve data preparation, model training, clustering labels, sending clusters to marketing teams, and deploying the model as a web portal for collaboration.
![Over 2000 Native and
Embedded Nodes are available
Nodes to Access Data from
➔ Databases [MySql, MongoDB]
➔ Files [CSV, txt, pdf, excel, PMML]
➔ Web, Cloud [REST, Web
Services]
Big Data Support
➔ Spark, HDFS
➔ Hive
➔ Impala
➔ In-database processing
Nodes to Transform Data
➔ Preprocessing data on the basis
of row, column, matrix
➔ Data Blending and Aggregation
➔ Feature Creation and Selection](https://image.slidesharecdn.com/knolxcreatingacustomersegmentationworkflowwithknime-200915093544/85/Creating-a-customer-segmentation-workflow-with-knime-10-320.jpg)
![Over 2000 Native and
Embedded Nodes are available
Nodes for Data Mining and Analysis
➔ Regression & Classification [Linear, Logistic, Decision
tree, SVM]
➔ Clustering [k-means, DBSCAN]
➔ Validation [cross-validation, scoring, ROC]
➔ Deep Learning [Keras, DL4J]
➔ External [R, Python, Weka]
Visualization
➔ Javascript based nodes [Scatter
plot, Box plot,ROC curve]
➔ Misc [Tag cloud, Open Street
Map]
➔ Script based visualizations [R,
Python
Deployment
➔ Writing results to Files [Excel,
CSV, Remote Storage]
➔ Writing results to Databases
➔ BIRT reporting](https://image.slidesharecdn.com/knolxcreatingacustomersegmentationworkflowwithknime-200915093544/85/Creating-a-customer-segmentation-workflow-with-knime-11-320.jpg)
Creating a customer segmentation workflow with knime
- 1. Presented By: Shubham Goyal Swantika Gupta Data Scientist Software Consultant Knoldus Inc. Knoldus Inc. Creating a Customer Segmentation Workflow with Knime
- 2. Lack of etiquette and manners is a huge turn off. KnolX Etiquettes Punctuality Respect Knolx session timings, you are requested not to join sessions after a 5 minutes threshold post the session start time. Feedback Make sure to submit a constructive feedback for all sessions as it is very helpful for the presenter. Silent Mode Keep your screens on mute unless you have a query Avoid Disturbance Avoid unwanted chit chat during the session.
- 3. Segment 1 Segment 2
- 4. Agenda Introduction to Knime Software What is customer segmentation Benefit and advantages Customer segmentation with Knime
- 5. ● Free and Open Source platform that provides capability for: ■ Data Analytics ■ Reporting ■ Integration ● Allows to Create and Productionize Data science using one easy and intuitive environment. ● Enables every stakeholder to focus on what he/she does best. Introduction to Knime
- 7. Knime Analytics Platform ● Open-source software for creating Data science processes. ● Strong and Comprehensive platform for ■ Drag-and-Drop Analytics ■ Machine Learning, Deep Learning, Natural Language Processing ■ ETL ■ Statistics
- 10. Over 2000 Native and Embedded Nodes are available Nodes to Access Data from ➔ Databases [MySql, MongoDB] ➔ Files [CSV, txt, pdf, excel, PMML] ➔ Web, Cloud [REST, Web Services] Big Data Support ➔ Spark, HDFS ➔ Hive ➔ Impala ➔ In-database processing Nodes to Transform Data ➔ Preprocessing data on the basis of row, column, matrix ➔ Data Blending and Aggregation ➔ Feature Creation and Selection
- 11. Over 2000 Native and Embedded Nodes are available Nodes for Data Mining and Analysis ➔ Regression & Classification [Linear, Logistic, Decision tree, SVM] ➔ Clustering [k-means, DBSCAN] ➔ Validation [cross-validation, scoring, ROC] ➔ Deep Learning [Keras, DL4J] ➔ External [R, Python, Weka] Visualization ➔ Javascript based nodes [Scatter plot, Box plot,ROC curve] ➔ Misc [Tag cloud, Open Street Map] ➔ Script based visualizations [R, Python Deployment ➔ Writing results to Files [Excel, CSV, Remote Storage] ➔ Writing results to Databases ➔ BIRT reporting
- 12. Knime Server ● Enterprise software for ○ Team-based collaboration ○ Automation of workflows ○ Management of workflows ○ Deployment of workflows as analytical applications and services
- 13. Knime Server
- 14. Advantages of using knime as solution Low Code Modular Scalable Plugin Based Inbuilt Collaboration
- 17. Can you imagine a writer, speaker or film director who never anticipated their audience?
- 19. Customer segmentation is the practice of dividing a customer base into groups of individuals that are similar in specific ways relevant to marketing, such as age, gender, interests and spending habits.
- 22. Data Collection Deciding what data will be collected and how it will be gathered Data Analyses Developing methods of data analysis for segmentation Segmentation Establishing effective communication among relevant business units (such as marketing and customer service) about the segmentation Implement Implementing applications to effectively deal with the data and respond to the information it provides 01 02 03 04 Customer Segmentation Procedure
- 23. K-means clustering ❏ K-means clustering is one of the simplest and popular unsupervised machine learning algorithms. ❏ The mean in k-means refers to averaging of the data: that is , finding the centroids in those data points. A cluster refers to a collection of data points aggregated together because of certain similarities.
- 24. How the K-means algorithm works To process the learning data, the K-means algorithm in data mining starts with a first group of randomly selected centroids, which are used as the beginning points for every cluster, and then performs iterative (repetitive) calculations to optimize the positions of the centroids It halts creating and optimizing clusters when either: ❏ The centroids have stabilized — there is no change in their values because the clustering has been successful. ❏ The defined number of iterations has been achieved.
- 26. Send those clusters to marketing or sales team Data preparation In this workflow, we will prepare data to feed into machine learning model Model deployment In this step we will deploy the model as a Web portal on knime server Collaboration Clusters labeling Model will give different clusters, we have to tag those clusters Model training In this workflow, we will train a machine learning model and save it Workflow Customer segmentation workflow with knime Knime Analytics Platform Knime server
- 27. Demo
- 28. Customer Segmentation for Domino’s ● Geographic ● Region – Domino’s outlet in different countries is a way of segmenting their market ● Potential Markets of different Regions ● City – They also segment the cities as class i, class ii, metros, small towns ● Demographic ● Age – under 13 years, 13 to 21 years, 21 to 35 years, 35 to 50 years, 50+ years. ● Family Income – Lower Middle Class, Middle Class, Upper Middle Class, High Class
- 30. ❏ Ability to Personalize Communication ❏ Upselling / Cross-selling Opportunities ❏ Higher ROI and CRO Benefits of Customer Segmentation
- 31. Thank You ! Get in touch with us: Lorem Studio, Lord Building D4456, LA, USA