Estadísticas y predicciones de LNB

Club Atlético San Martín

San Martín ha mostrado una consistencia notable esta temporada, gracias a su sólida defensa y un ataque bien coordinado. La clave de su éxito ha sido su capacidad para mantener la calma bajo presión y ejecutar jugadas estratégicas en momentos críticos.

Gimnasia y Esgrima

Gimnasia y Esgrima ha impresionado con su velocidad y agilidad en el campo. Su habilidad para realizar transiciones rápidas del ataque a la defensa les ha dado una ventaja significativa en muchos partidos.

Quimsa

Conocido por su tenacidad y espíritu competitivo, Quimsa no deja piedra sin remover. Su juego colectivo y la experiencia de sus jugadores clave les han permitido mantenerse en la cima durante gran parte de la temporada.

Boca Juniors

Boca Juniors es un equipo que siempre aspira a lo más alto. Con una mezcla de juventud e inteligencia táctica, han logrado sorprender a muchos rivales esta temporada.

Club Atlético San Martín vs. Gimnasia y Esgrima

Predicción: Victoria ajustada para San Martín. Su defensa sólida podría ser suficiente para contener el ataque rápido de Gimnasia y Esgrima.

Quimsa vs. Peñarol

Predicción: Victoria para Quimsa. Su experiencia y cohesión como equipo les darán una ventaja sobre Peñarol.

Boca Juniors vs. Regatas Corrientes

Predicción: Empate o victoria ajustada para Boca Juniors. La presión será alta, pero su juventud e inteligencia táctica podrían llevarlos a la victoria.

San Lorenzo vs. Instituto

Predicción: Victoria cómoda para San Lorenzo. Su dominio local y experiencia deberían ser suficientes para superar a Instituto.

Tips Adicionales para Apuestas

• Fíjate en las estadísticas recientes de los jugadores clave antes de apostar.
• Ten en cuenta las condiciones del campo y cualquier posible ausencia debido a lesiones.
• Observa las tendencias recientes de cada equipo para identificar patrones que puedan influir en el resultado del partido.
• No te olvides de diversificar tus apuestas para minimizar riesgos.

Análisis Técnico-Defensivo

Estudia cómo cada equipo gestiona su defensa durante los últimos minutos del partido. Los equipos que mejor gestionan el tiempo pueden tener una ventaja crucial.

Evaluación Psicológica del Equipo

Considera el estado psicológico del equipo antes del partido. Los equipos con buena moral tras una racha ganadora pueden tener un impulso adicional.

Análisis Estadístico Detallado

Utiliza herramientas estadísticas avanzadas para analizar datos históricos y predecir resultados con mayor precisión.

Gestión del Bankroll

Mantén un control estricto sobre tu presupuesto de apuestas para evitar pérdidas significativas.

Cómo Implementar Estas Estrategias

1. Recolección de Datos: Recoge datos detallados sobre cada equipo y jugador antes del partido.
2. Análisis Comparativo: Compara las estadísticas recientes con el rendimiento histórico para identificar tendencias.
3. Simulación: Realiza simulaciones basadas en diferentes escenarios posibles durante el partido.
4. Toma de Decisiones Informada: Utiliza toda la información recopilada para tomar decisiones informadas sobre tus apuestas.

sagarshah0000/COMP90055<|file_sep|>/assignment1/README.md # COMP90055 - Assignment 1 ## Introduction This is the first assignment of the course COMP90055 - Data Science for Business Analytics at the University of Melbourne. The assignment consists of three parts: 1. **Linear Regression** - In this part you will build a linear regression model to predict the price of houses in King County using various features of the houses. 2. **Logistic Regression** - In this part you will build and evaluate logistic regression models to classify whether an individual will be diagnosed with diabetes or not. ## How to Run ### Prerequisites You need to have Python 3 and the following Python packages installed: * numpy * pandas * matplotlib * sklearn ### Running the code To run the code for Linear Regression and Logistic Regression separately: python linear_regression.py python logistic_regression.py To run both the parts together: python assignment1.py The code for each part can also be run separately by running the respective files. ### Running in Google Colab To run the code in Google Colab: 1. Open https://colab.research.google.com/ 2. Click on `New Notebook` 3. Go to `File` -> `Upload notebook` and select either `linear_regression.py` or `logistic_regression.py` depending on which part you want to run. 4. The first cell in the notebook should have `!pip install numpy pandas matplotlib sklearn` written in it. 5. Click on `Runtime` -> `Run all` 6. This should execute all the cells one by one and you will see the results at the end. <|repo_name|>sagarshah0000/COMP90055<|file_sep|>/assignment1/logistic_regression.py import numpy as np import pandas as pd import matplotlib.pyplot as plt from sklearn import metrics from sklearn.model_selection import train_test_split from sklearn.linear_model import LogisticRegression # Read data from file data = pd.read_csv('diabetes.csv') # Split data into X (features) and Y (target) X = data.iloc[:, 0:-1].values Y = data.iloc[:, -1].values # Split data into training set and test set X_train, X_test, Y_train, Y_test = train_test_split(X,Y,test_size=0.25) # Create logistic regression model and fit on training set model = LogisticRegression(solver='lbfgs') model.fit(X_train,Y_train) # Evaluate model on test set predictions = model.predict(X_test) accuracy = metrics.accuracy_score(Y_test,predictions) print('Accuracy:', accuracy) # Confusion matrix print('Confusion matrix:') print(metrics.confusion_matrix(Y_test,predictions)) # Precision recall curve Y_prob = model.predict_proba(X_test) Y_prob = Y_prob[:, 1] precision_curve, recall_curve, pr_thresholds = metrics.precision_recall_curve(Y_test,Y_prob) closest_zero = np.argmin(np.abs(pr_thresholds)) closest_zero_p = precision_curve[closest_zero] closest_zero_r = recall_curve[closest_zero] plt.plot(recall_curve, precision_curve,label='Precision-Recall Curve') plt.plot(closest_zero_r, closest_zero_p,'o', markersize=12, label='Threshold Zero', fillstyle='none', c='r', mew=3) plt.xlabel('Recall') plt.ylabel('Precision') plt.legend(loc=0) plt.show() # ROC curve fpr,tpr,thresholds = metrics.roc_curve(Y_test,Y_prob) auc_score = metrics.auc(fpr,tpr) plt.plot(fpr,tpr,label='ROC Curve (area = %0.2f)'%auc_score) plt.plot([0,1],[0,1],'r--') plt.xlim([0.0,1]) plt.ylim([0.0,1]) plt.xlabel('False Positive Rate') plt.ylabel('True Positive Rate') plt.title('Receiver operating characteristic') plt.legend(loc=4) plt.show() # Plot histograms of each feature grouped by outcome for i in range(8): col_name = data.columns[i] plt.hist(data[col_name][data['Outcome']==0],bins=50,alpha=0.5,color='blue',label='Non-diabetic') plt.hist(data[col_name][data['Outcome']==1],bins=50,alpha=0.5,color='red',label='Diabetic') plt.xlabel(col_name) plt.ylabel('Frequency') plt.legend(loc='upper right') plt.show() <|file_sep|># COMP90055 - Assignment 2 ## Introduction This is the second assignment of the course COMP90055 - Data Science for Business Analytics at the University of Melbourne. The assignment consists of two parts: 1. **Clustering** - In this part you will cluster customer data into groups using k-means clustering algorithm and then analyse these clusters to determine what characteristics distinguish each cluster from others. 2. **Association Rule Mining** - In this part you will use association rule mining algorithm to find out interesting relationships between products purchased by customers. ## How to Run ### Prerequisites You need to have Python 3 and the following Python packages installed: * numpy * pandas * matplotlib * seaborn * scikit-learn ### Running the code To run the code for Clustering and Association Rule Mining separately: python clustering.py python association_rule_mining.py To run both the parts together: python assignment2.py The code for each part can also be run separately by running the respective files. ### Running in Google Colab To run the code in Google Colab: 1. Open https://colab.research.google.com/ 2. Click on `New Notebook` 3. Go to `File` -> `Upload notebook` and select either `clustering.py` or `association_rule_mining.py` depending on which part you want to run. 4. The first cell in the notebook should have `!pip install numpy pandas matplotlib seaborn sklearn` written in it. 5. Click on `Runtime` -> `Run all` 6. This should execute all the cells one by one and you will see the results at the end. <|file_sep|># COMP90055 - Assignment 4 ## Introduction This is the fourth assignment of the course COMP90055 - Data Science for Business Analytics at University of Melbourne. The assignment consists of two parts: 1) **Recommendation Systems** - In this part you will build recommendation systems based on users' ratings of movies. ## How to Run ### Prerequisites You need to have Python 3 and the following Python packages installed: * numpy * pandas * matplotlib * scikit-learn ### Running the code To run recommendation systems: python recommendation_systems.py The code can also be run separately by running `recommendation_systems.py`. ### Running in Google Colab To run the code in Google Colab: 1) Open https://colab.research.google.com/ 2) Click on New Notebook. 3) Go to File -> Upload notebook and select recommendation_systems.py. 4) The first cell in notebook should have !pip install numpy pandas matplotlib sklearn written in it. 5) Click on Runtime -> Run all. 6) This should execute all cells one by one and you will see results at end. <|repo_name|>sagarshah0000/COMP90055<|file_sep|>/assignment4/recommendation_systems.py import numpy as np import pandas as pd import matplotlib.pyplot as plt def mean_absolute_percentage_error(y_true,y_pred): return np.mean(np.abs((y_true-y_pred)/y_true))*100 def get_movie_recommendations(user_id,movie_titles,user_ratings,ratings_matrix,num_recommendations): # Get index of user's ratings from ratings matrix user_index = np.where(ratings_matrix.index == user_id)[0][0] # Get user's predicted ratings for all movies by multiplying user vector with movie vectors from ratings matrix user_pred_ratings = ratings_matrix.iloc[user_index,:].values @ ratings_matrix.T.values # Get user's actual ratings for movies they have already rated from user_ratings dataframe user_actual_ratings = user_ratings[user_ratings['user_id'] == user_id]['rating'].values[0] # Create dataframe with movie ids and corresponding predicted ratings for user movie_ids = ratings_matrix.columns.to_list() pred_ratings_df = pd.DataFrame({'movie_id':movie_ids,'pred_rating':user_pred_ratings}) # Remove movies that user has already rated from prediction dataframe pred_ratings_df.drop(pred_ratings_df[pred_ratings_df['movie_id'].isin(user_ratings[user_ratings['user_id'] == user_id]['movie_id'])].index,inplace=True) # Sort dataframe based on predicted rating value so that highest predicted rating is at top pred_ratings_df.sort_values(by=['pred_rating'],ascending=False,inplace=True) # Get movie titles corresponding to top num_recommendations movies from movie_titles dataframe top_movies_indices = pred_ratings_df.head(num_recommendations)['movie_id'].to_list() top_movie_titles = [] for index in top_movies_indices: top_movie_titles.append(movie_titles[movie_titles['movieId'] == index]['title'].values[0]) return top_movie_titles # Read data from files into dataframes movies_df = pd.read_csv('movies.csv') ratings_df = pd.read_csv('ratings.csv') # Group ratings dataframe by movie id and get average rating for each movie rounded off to nearest integer value avg_movie_rating_df = pd.DataFrame(ratings