Data science process: 6. Present Findings and Build Application

Data science process is a series of activities that help us move from a business problem to a useful solution based on data. 1. Define research goals 2. Data retrieval 3. Data preparation 4. Data exploration 5. Model building 6. Presenting results and automation

6. Present Findings and Build Application

•  After completion of data analysis and building a well performing model the next step is to ‒

1) Present findings clearly to teams, managements or clients.

2) Automate data analysis so the model can be regularly used.

Importance of presentation and automation

1) The value of data science comes not just by building models but from how well the project is communicated and applied for real‒life problem solving.

2) If people understand the findings from your data analysis, they can make better business decisions.

3) Automation of data analysis ensures to stay updated without repeating the same work manually.

1. Present Your Findings

•  We need to present the findings of the data analysis in order to communicate the insights, predictions and recommendations in simple and meaningful way to everyone.

•  For presenting the findings to the technical persons you need to discuss methods, accuracy and assumptions.

•  Replace the numbers with clear charts and dashboards. For example ‒ A bar chart showing "Sales growth by region" is easier to understand than using long table.

 •   Make use of some common presentation tools. These tools and their purpose are given in the following table

Presentation tool ‒  Purpose

Powerpoint or Google Slides :‒ Present reports and summaries.

Excel or Google Sheets  :‒ Shares data summaries and charts  

Power BI :‒ Creates interactive dashboards.

Jupyter Notebook :‒ Combine Code and results.

2. Build the Application and Automate

•  After building the models, we want to use them regularly. For example ‒ we need to generate daily reports or update predictions automatically.

• Common tools for automation are Python, Excel, Power BI, or Application Programming Interfaces(APIs).

•  For example ‒ If we build a model "Customer Churn(leaving service) Prediction Model for a telecom company".

1) We present our result: The model predicts 90% accuracy in identifying customers likely to leave. Visualize results in a PowerPoint or dashboard.

2) We automate it: Every month, new customer data is fed into the model. The output updates an online dashboard automatically for the marketing team.

•  This saves time and ensures continuous insights. Ultimately it helps in overall improvement in taking business decisions.

Role of soft skills

• The final stage of data science process that is ‒ "Presenting results and Automation" requires communication, presentation and teamwork.

• Data scientists must:

1.Explain complex results in simple language,

2.Work with managers and decision‒makers to turn insights into action.

3.Collaborate with developers or IT teams to deploy models.

Review Questions

1.What is the purpose of the "Present Findings" stage in the data science process?

2. Why is it important to present your findings clearly after building a model?

3.What is meant by model automation, explain with example. Why is it necessary?

4.Give examples of soft skills that are helpful when presenting data science results.