Mortgage rates surged at an unprecedented pace in 2022, with the average 10-year fixed rate doubling between March and December of that year. In response to mounting inflation, the Bank of England implemented a series of rate hikes, pushing borrowing costs steadily higher. By October 2025, the average 10-year fixed mortgage rate stood at **** percent. As financing becomes more expensive, housing demand has cooled, weighing on market sentiment and slowing house price growth. How have the mortgage hikes affected the market? After surging in 2021, the number of residential properties sold fell significantly in 2023, dipping to just above *** million transactions. This contraction in activity also dampened mortgage lending. Between the first quarter of 2023 and the first quarter of 2024, the value of new mortgage loans declined year-on-year for five consecutive quarters. Even as rates eased modestly in 2024 and housing activity picked up slightly, volumes remained well below the highs recorded in 2021. How are higher mortgages impacting homebuyers? For homeowners, the impact is being felt most acutely as fixed-rate deals expire. Mortgage terms in the UK typically range from two to ten years, and many borrowers who locked in historically low rates are now facing significantly higher repayments when refinancing. By the end of 2026, an estimated five million homeowners will see their mortgage deals expire. Roughly two million of these loans are projected to experience a monthly payment increase of up to *** British pounds by 2026, putting additional pressure on household budgets and constraining affordability across the market.

Summary The proposed competition challenges professionals to analyze a custom dataset of Italian companies that went bankrupt in 2023. The dataset contains company IDs categorized by industry:

Axxxxx: Commerce Bxxxxx: Online Services Cxxxxx: Hospitality Dxxxxx: Construction Exxxxx: Restaurants Fxxxxx: Industry Gxxxxx: Startups The primary goal is to utilize machine learning algorithms to:

Identify Key Variables: Determine which features in the dataset significantly influence bankruptcy. Class-Based Analysis: Investigate whether different variables affect bankruptcy across company categories (ID groups). Predict Bankruptcy: Build models to predict the likelihood of bankruptcy based on the data. Handle Multicollinearity: Identify and mitigate issues of multicollinearity among variables. Evaluate Models: Use performance metrics like confusion matrices, ROC curves, AUC scores, and other evaluation criteria. Minimize False Positives: Achieve a balance between detection rate and false alarm rate while ensuring the models remain reliable, fast, and interpretable. Machine Learning Techniques Participants are required to apply and compare multiple machine learning algorithms, including:

Random Forest Support Vector Machines (SVM) Neural Networks Linear and Logistic Regression Models must be trained and tested on appropriate data splits, with comprehensive analysis performed on both training and test datasets.

Analysis This task involves several critical steps in machine learning analysis and evaluation:

Data Preparation and Feature Selection

Perform feature selection to reduce dimensionality and enhance model interpretability. Consider techniques to remove redundant or irrelevant features while retaining the most informative variables. Model Evaluation Metrics Participants will evaluate models using multiple performance metrics:

Predictive Accuracy: Ratio of correct predictions to total cases. True Positive Rate (Recall/Sensitivity): Ability of the model to detect bankrupt companies correctly. False Positive Rate (Fall-Out): Percentage of companies incorrectly classified as bankrupt. Precision: Proportion of true positives among all predicted positives. F1-Score: Weighted harmonic mean of precision and recall. Confusion Matrix: Visualization of classification performance. ROC Curve and AUC: Trade-offs between detection rate and false alarm rate. Class-Specific Analysis

Explore whether distinct variables influence bankruptcy differently across industry categories (e.g., financial variables for commerce vs. operational metrics for startups). Multicollinearity

Address multicollinearity using techniques like Variance Inflation Factor (VIF) to ensure that correlated predictors do not negatively impact model performance. Trade-Off Between Performance and Practicality

Balance high accuracy and low false positive rates with the need for computational efficiency and interpretability. Key Objectives for Participants Develop robust and scalable models to predict bankruptcy effectively. Identify critical variables for each industry group and assess their influence. Provide actionable insights through clear and concise visualizations and reports. Optimize performance to achieve reliable predictions while minimizing false alarms. Demonstrate expertise in applying machine learning algorithms and interpreting their results. This competition not only tests the technical capabilities of participants but also their ability to derive meaningful business insights from data, making it a compelling challenge for data science professionals.

The dataset is designed to assist in predicting recommended medications for patients based on their fever condition, symptoms, medical history, and other relevant factors. It incorporates a mix of patient health data, environmental variables, and lifestyle choices to improve model accuracy and better simulate real-world scenarios.

Dataset Characteristics: Total Samples: 1000 (modifiable based on user needs). Number of Features: 19 features + 1 target column. File Format: CSV (enhanced_fever_medicine_recommendation.csv). Features Description: Column Name Description Data Type Temperature Body temperature of the patient in Celsius (e.g., 36.5 - 40.0). Float Fever_Severity Categorized fever severity: Normal, Mild Fever, High Fever. Categorical Age Age of the patient (1-100 years). Integer Gender Gender of the patient: Male or Female. Categorical BMI Body Mass Index of the patient (e.g., 18.0 - 35.0). Float Headache Whether the patient has a headache: Yes or No. Categorical Body_Ache Whether the patient has body aches: Yes or No. Categorical Fatigue Whether the patient feels fatigued: Yes or No. Categorical Chronic_Conditions If the patient has any chronic conditions (e.g., diabetes, asthma): Yes or No. Categorical Allergies If the patient has any allergies to medications: Yes or No. Categorical Smoking_History If the patient has a history of smoking: Yes or No. Categorical Alcohol_Consumption If the patient consumes alcohol: Yes or No. Categorical Humidity Current humidity level in the patient’s area (e.g., 30-90%). Float AQI Current Air Quality Index in the patient’s area (e.g., 0-500). Integer Physical_Activity Daily physical activity level: Sedentary, Moderate, Active. Categorical Diet_Type Diet preference: Vegetarian, Non-Vegetarian, or Vegan. Categorical Heart_Rate Resting heart rate of the patient in beats per minute (e.g., 60-100). Integer Blood_Pressure Blood pressure category: Normal, High, or Low. Categorical Previous_Medication Medication previously taken by the patient: Paracetamol, Ibuprofen, Aspirin, or None. Categorical Recommended_Medication Target variable indicating the recommended medicine: Paracetamol or Ibuprofen. Categorical

Problem Statement

The data is related with direct marketing campaigns (phone calls) of a Portuguese banking institution. The classification goal is to predict if the client will subscribe a term deposit.

File Format -

.csv file with descriptions as given below -

Content

Input variables: 1 - age (numeric) 2 - job : type of job (categorical: 'admin.','blue-collar','entrepreneur','housemaid','management','retired','self-employed','services','student','technician','unemployed','unknown') 3 - marital : marital status (categorical: 'divorced','married','single','unknown'; note: 'divorced' means divorced or widowed) 4 - education (categorical: 'basic.4y','basic.6y','basic.9y','high.school','illiterate','professional.course','university.degree','unknown') 5 - default: has credit in default? (categorical: 'no','yes','unknown') 6 - housing: has housing loan? (categorical: 'no','yes','unknown') 7 - loan: has personal loan? (categorical: 'no','yes','unknown') related with the last contact of the current campaign: 8 - contact: contact communication type (categorical: 'cellular','telephone') 9 - month: last contact month of year (categorical: 'jan', 'feb', 'mar', ..., 'nov', 'dec') 10 - day_of_week: last contact day of the week (categorical: 'mon','tue','wed','thu','fri') 11 - duration: last contact duration, in seconds (numeric). Important note: this attribute highly affects the output target (e.g., if duration=0 then y='no'). Yet, the duration is not known before a call is performed. Also, after the end of the call y is obviously known. Thus, this input should only be included for benchmark purposes and should be discarded if the intention is to have a realistic predictive model.

other attributes:

12 - campaign: number of contacts performed during this campaign and for this client (numeric, includes last contact) 13 - pdays: number of days that passed by after the client was last contacted from a previous campaign (numeric; 999 means client was not previously contacted) 14 - previous: number of contacts performed before this campaign and for this client (numeric) 15 - poutcome: outcome of the previous marketing campaign (categorical: 'failure','nonexistent','success')

social and economic context attributes

16 - emp.var.rate: employment variation rate - quarterly indicator (numeric) 17 - cons.price.idx: consumer price index - monthly indicator (numeric) 18 - cons.conf.idx: consumer confidence index - monthly indicator (numeric) 19 - euribor3m: euribor 3 month rate - daily indicator (numeric) 20 - nr.employed: number of employees - quarterly indicator (numeric)

Output variable (desired target): 21 - term_deposit - has the client subscribed a term deposit? (binary: 'yes','no')

Acknowledgements

Source:

[Moro et al., 2014] S. Moro, P. Cortez and P. Rita. A Data-Driven Approach to Predict the Success of Bank Telemarketing. Decision Support Systems, Elsevier, 62:22-31, June 2014

This dataset contains the predicted prices of the asset NotifAi News over the next 16 years. This data is calculated initially using a default 5 percent annual growth rate, and after page load, it features a sliding scale component where the user can then further adjust the growth rate to their own positive or negative projections. The maximum positive adjustable growth rate is 100 percent, and the minimum adjustable growth rate is -100 percent.

The "Stock Market Dataset for AI-Driven Prediction and Trading Strategy Optimization" is designed to simulate real-world stock market data for training and evaluating machine learning models. This dataset includes a combination of technical indicators, market metrics, sentiment scores, and macroeconomic factors, providing a comprehensive foundation for developing and testing AI models for stock price prediction and trading strategy optimization.

Key Features Market Metrics:

Open, High, Low, Close Prices: Daily stock price movement. Volume: Represents the trading activity during the day. Technical Indicators:

RSI (Relative Strength Index): A momentum oscillator to measure the speed and change of price movements. MACD (Moving Average Convergence Divergence): An indicator to reveal changes in strength, direction, momentum, and duration of a trend. Bollinger Bands: Upper and lower bands around a stock price to measure volatility. Sentiment Analysis:

Sentiment Score: Simulated sentiment derived from financial news and social media, ranging from -1 (negative) to 1 (positive). Macroeconomic Factors:

GDP Growth: Indicates the overall health and growth of the economy. Inflation Rate: Reflects changes in purchasing power and economic stability. Target Variable:

Buy/Sell Signal: Binary classification (1 = Buy, 0 = Sell) based on price movement thresholds, simulating actionable trading decisions. Use Cases AI Model Training: Ideal for building stock prediction models using LSTM, Gradient Boosting, Random Forest, etc. Trading Strategy Optimization: Enables testing of trading algorithms and strategies in a simulated environment. Sentiment Analysis Research: Useful for understanding how sentiment influences stock movements. Feature Engineering and Selection: Provides a diverse set of features for experimentation with advanced techniques like PCA and LDA. Dataset Highlights Synthetic Yet Realistic: Carefully designed to mimic real-world financial data trends and relationships. Comprehensive Coverage: Includes key indicators and metrics used by traders and analysts. Scalable: Suitable for use in both small-scale academic projects and larger AI-driven trading platforms. Accessible for All Levels: The intuitive structure ensures that even beginners can utilize this dataset for financial machine learning applications. File Format The dataset is provided in CSV format, where:

Rows represent individual trading days. Columns represent features (technical indicators, market metrics, etc.) and the target variable. Acknowledgments This dataset is synthetically generated and is intended for research and educational purposes. It is not based on real market data and should not be used for actual trading.

This dataset contains the predicted prices of the asset Buy The News over the next 16 years. This data is calculated initially using a default 5 percent annual growth rate, and after page load, it features a sliding scale component where the user can then further adjust the growth rate to their own positive or negative projections. The maximum positive adjustable growth rate is 100 percent, and the minimum adjustable growth rate is -100 percent.

The benchmark interest rate in Portugal was last recorded at 2.15 percent. This dataset provides - Portugal Interest Rate - actual values, historical data, forecast, chart, statistics, economic calendar and news.

Mortgage Rate in the United Kingdom remained unchanged at 6.78 percent in October. This dataset provides - United Kingdom BBA Mortgage Rate- actual values, historical data, forecast, chart, statistics, economic calendar and news.

The benchmark interest rate in Ireland was last recorded at 4.50 percent. This dataset provides - Ireland Interest Rate - actual values, historical data, forecast, chart, statistics, economic calendar and news.

Mortgage Rate in Australia decreased to 5.51 percent in September from 5.52 percent in August of 2025. This dataset includes a chart with historical data for Australia Mortgage Rate.