Kernix supports you throughout your data project with a 4-step process:
- Business understanding: understanding the business objectives in order to translate them into data-driven use cases and to define project performance measures.
- Identification and preparation of relevant data for the project, whether from internal or external sources.
- Analysis and modelling of the data using techniques such as machine learning, text mining, graph mining and others.
- Demonstration of value creation through data visualization, an online application or simply scripts to implement the models.
Describe use cases and performance measures.
- Define business objectives
- Assess the current situation (resources, assumptions, constraints…)
- Define the objectives in terms of data mining
- Establish a project plan
Analyse the true potential of data and choose the most promising use cases.
- Identify, collect and enrich data
- Explore and visualise data
- Merging different sources
- Select and clean relevant data
- Matching identified opportunities with key success factors
Analysis and modelling
Realising and modelling creation opportunities.
- Adopt a modelling technique (Machine Learning, Deep learning, Semantic analysis, Data mining)
- Develop the models
- Analyse and optimise the algorithms
- Validate the results
Develop the tools to convince and go further. Industrialise or automate value creation. Explore other use cases for other opportunities.
- Create an application that fulfils the key success factors
- Present the results
- Demonstrate the value created
- Consider industrialisation together