Data Mining Applications & Real life case study

If there is one thing that I would change about my data science and journey of getting to software engineer and data engineer, that would be bravery and just get into it. There are types of data mining and text mining. I used logistic regression for my hypothesis-driven thesis on 3PL (Third-party logistics) undergraduate thesis. Then, I worked as a digital marketing, and this background undermine the tasks of discovery-driven and lead insights.

Learning about the patterns:

1. Consumer behaviour
2. Resilience and persevere
3. Data management — ETL process

4. Customer Communication in Marketing

Data science can produce information about real-time events and allow marketers to tap into those situations to target customers. For example, marketers of a hotel company can use data science in real-time to determine travelers whose flights were delayed. They can then target them by sending ad campaigns directly to their mobile devices.

Providing a rich customer experience has always been an important factor in achieving marketing success. With data science, marketers can collect user behavior patterns that will predict who may want or need specific products. This allows them to market efficiently and provide customers with enriching experiences.

5. What is AI? Cause definitely, this term is overrated!

In this module, we have learned about issues and applications of Data warehousing technologies — the kinds of issues that you will encounter in industry.

Make sure you understand and can apply the principles/steps of:

• Data warehouse: Snowflake and Star Schemas
• Data Cubes

Now that you’re at the end of this module, it’s time to review and reflect on what you’ve learnt.

Reflection

Reflect on what you’ve learnt. Here are some pointers to get you started.

• Describe in your own words what you’ve learnt.
• Why is it important to learn? How does it connect to what you already know? Was it difficult? Why/why not?
• What are some examples of this in action? How can you put your learning into practice?

Still unsure?

Look back at the pages in this post and check the discussions. As always, if you have any questions that others might also ask, you can post in the discussions. If you are still confused, contact me at noviaayup@live.com directly.

How to start a good #MachineLearning project?

At a high-level:

1. Learn Machine Learning

YouTube-

LinkedIn Learning

2. Find a real-world problem

Look for a problem that interests you (Healthcare, Ecommerce, etc..)

List of Datasets: https://lnkd.in/gTft-GV

3. Exploratory Data Analysis (Python or R)

Understand what patterns and values your data has. Apply different visualizations and statistical testings to back up our findings. Go out and explore!

4. Use Multiple Models (Simple VS Complex)

Try the simplest models first. You’ll be amazed how far simple models like Linear/Logistic Regression will take you. Once you’ve narrowed down and understand how your model performs, then try out more complex models (Deep Learning).

5. Document your work (GitHub or Blog)

This is the most important. Document all of your analysis, visualizations, and experiments. Document your failures. Understand why it worked and why it didn’t worked.

6. Showcase your work

Share your project with others. Communicate your results & get feedback from the community. Be proud! This allows you to reflect back on some mistakes you’ve made and gather further insights to improve on.

#github, #skillshare, #angellist, #ASKHN:HiringNow,