- Machine Learning models are ways to automatically describe or predict real-world phenomena.
- Descriptive models try to describe things. Those things are sometimes groups or segments in your data, or sometimes processes that underly your data.
- A classic example is customer segmentation. For example, think about grouping customers from your customer database to be able to send only relevant publicity to your customers. In this link you'll find an example of customer segmentation that I've worked on.
- Models that explain processes are not focused on grouping the data, but they are focused on finding out what is going on. An example that I've worked on is modeling a cheese production line to find out which are the factors that influence quality of the production line. Or similary I have worked on models that try to explain quality of fruit and vegetables by modeling hte relation between quality and other factors.
- Predictive models on the other hand try to predict or forecast something that will happen in the future. An example that I have worked on is the merchandise sales of Disneyland. Those models are different in the way that you can verify aftwerwards how accurate the forecast has been.
- I'm always open for a chat! Just send me a message over here. Don't worry if you don't know what you need, I also help with Data Strategy
- Data is a valuable resource, but only when you know how to use it.
- Data Mining - identifying relationships between different data sources - can help to identify what data is valuable and what data contains more noise than signal. This can be of much use when developing a data strategy.
- Another example of Data Strategy related work is the development and implementation of reporting and KPIs (Key Performance Indicators). In my previous online marketing projects, this has always played an important role.
- By the way, if you prefer learning how to do the work yourself, I also do training and education sessions. I'm always open for a chat! Just send me a message over here.
Training and Education
- Whether it is to transfer skills to a technical team or to share the possibilities of data science with skeptics, training and education sessions are always an interesting way to broaden your horizons.
- Technical hands-on sessions allow people to have a quick start in data science technologies like R, R Shiny, Python, or others.
- Sharing sessions with relatively unexperienced teams are a great way to get people started to reflect on possibilities of data science and to get up to speed with teh current developments.
- Teaching people methodology for research can help a lot in deepening understanding of R&D work.
- Has this sparked your interest? Just send me a message over here.