A.Overview of the Project:

In my role as a Business Intelligence Analyst at ESN Data, I embarked on a transformative mission with "Les plus Beaux Logis de Paris," a prominent client in the real estate sector. This journey involved delving into the historical tapestry of real estate prices in Paris, weaving together insights that would shape strategic decision-making through the power of predictive analysis.

Upon my return to the office, I encountered Jade, the director of the service, displaying a sense of concern. A comprehensive analysis revealed two pivotal dimensions for my predictive algorithm:

• The predictive algorithm's regression performance needed to boast an error rate below 10%, ensuring accurate forecasts of real estate prices.
• The predictive insights would serve as a compass, guiding strategic decisions towards the most valuable segments of the portfolio.

These insights were not just numerical; they were stories waiting to be told. Soft skills, such as effective communication and reassurance, became the threads that would weave these stories into actionable narratives for the client. Jade acknowledged and valued the quality of my work, extending an invitation for collaborative exploration of statistical techniques. The client faced a pressing challenge: swift decision-making in real estate acquisitions amidst limited property information. Opportunities lay hidden, requiring manual investigation. Jade envisioned a future where this process could be automated, seeking my expertise to illuminate the path forward.

B.Steps:

1.Analyzing Historical Data:

-Goal:Unearth the tales hidden within past real estate trends in Paris.

-Steps:Explore data nuances, from types and categories to transaction volumes, historical intervals, and price variations.

2.Training the Prediction Algorithm:

-Goal:Craft a model that breathes life into property value predictions.

-Steps:Dance with linear regression, aiming for an error less than 10%. Engage in a critical assessment of the model's relevance.

-Keep in Mind:Let one-hot encoding and train-test splitting be your dance partners, consulting the Scikit-Learn documentation as your trusted choreographer.

3.Future Portfolio Valuation Prediction:

-Goal:Cast a vision of the future portfolio's financial symphony and present the results as a harmonious narrative.

-Steps:Polish the code, predict the future valuation, and present findings in a non-technical manner.

4.Unsupervised Property Classification:

-Goal:Bring forth an automated classification ballet for property opportunities.

5.Analysis and Results Presentation:

-Goal:Conduct a symphony of analysis and present the results in a language comprehensible to all.

-Steps:Dive into the results, creating a non-technical presentation that resonates with the client.

C.Resources:

• Machine Learnia(In French)

• K-means expliqué: exemple pratique en Python (In French)

• Complete Python Pandas Data Science Tutorial! (Reading CSV/Excel files, Sorting, Filtering, Groupby)

  SKILLS:

  HARD SKILLS:

-Data Analysis with Pandas:The project involved extensive data analysis using Pandas, requiring a deep understanding of data manipulation, exploration, and interpretatio

-Machine Learning - Linear Regression:Employing linear regression for predictive modeling demanded expertise in machine learning concepts. It involved understanding the algorithm, feature engineering, and model evaluation.

-Python Programming:The entire project was executed using Python, necessitating proficiency in the programming language. This included coding for data analysis, machine learning, and presentation preparation.

-K-means Clustering:Implementing the K-means algorithm for unsupervised property classification required knowledge of clustering techniques and their application to real-world datasets.

SOFT SKILLS:

-Communication:Clear communication was crucial throughout the project. From interacting with the client to presenting complex findings, I honed my ability to convey technical information in a non-technical and reassuring manner.

- Critical Thinking: Addressing challenges such as the need for automation demanded a critical evaluation of the available data, methodologies, and potential limitations. It required me to think beyond immediate solutions and anticipate future needs.

- Adaptability:Navigating through different project phases, from data analysis to predictive modeling and client presentations, required adaptability. It involved adjusting strategies based on feedback, challenges, and evolving project requirements.

REFERENCES:

• Initiate yourself to Python for data analysis.(In French)
• Create a slideshow to accompany your presentation(In French)
• Initiate yourself to Machine Learning(In French)