Estadísticas y predicciones de A1 Liga

A través del análisis meticuloso utilizando estas métricas avanzadas se pueden hacer predicciones mucho más precisas sobre posibles resultados y desempeños destacados dentro del juego.

Predicciones Basadas en Estadísticas Avanzadas:

• Cibona Zagreb vs Cedevita Olimpija: Dada la alta eficiencia ofensiva observada recientemente por parte de Cedevita combinada con el índice +/- neto positivo cuando sus estrellas están activas simultáneamente; aún así debería prestar atención al sólido rendimiento +/- neto mostrado por Cibona cuando sus principales defensores están presentes.
• johnnynorth/Intelligent-System-Project<|file_sep|>/README.md # Intelligent-System-Project ## Goal This is the project for Intelligent System Course in the National University of Singapore. The goal is to build an intelligent system that helps people who are in need to find nearby resources using their smartphones. The system is based on an online crowdsourcing platform called Amazon Mechanical Turk (MTurk). It has two main components: 1) MTurk task to collect data about the resources (i.e., places and things) from people who know the area. 2) Machine learning algorithms to cluster the collected data into categories and then provide recommendation. ## Data collection We used MTurk to collect data about the resources in Singapore from people who know the area. The task was designed as follows: * Show participants an image of a map and ask them to mark on it where they think there are resources for people who are in need. * Ask participants to describe the resources in their own words. * Ask participants to provide additional information such as the type of resource (e.g., food bank), location (e.g., address), opening hours, etc. * Ask participants to rate how important they think each resource is on a scale from 1 to 10. We collected data from over **1000** participants and ended up with more than **6000** resource entries. ## Data preprocessing The collected data was preprocessed using natural language processing techniques such as tokenization, stemming/lemmatization and stop word removal. We also performed entity recognition using the Stanford Named Entity Recognizer (NER) to extract location names from the text descriptions provided by participants. Finally we normalized the opening hours information by converting it into a standard format such as "Monday-Friday:9am-6pm". ## Clustering algorithm We used K-means clustering algorithm to group similar resources together based on their text descriptions and other features such as location and opening hours. We experimented with different numbers of clusters and evaluated the results using silhouette score metric which measures how well-separated each cluster is from other clusters. The best results were obtained with **10** clusters which had an average silhouette score of **0.65**. ## Recommendation system We built a simple recommendation system that takes as input the current location of the user and outputs a list of nearby resources sorted by their importance rating given by participants. To do this we first filtered out all resources that are not within **500m** radius from the user's location using Haversine distance formula which calculates distance between two points on Earth's surface given their latitude/longitude coordinates. Then we sorted the remaining resources by their importance rating and returned top **10** results. <|file_sep|># -*- coding: utf-8 -*- """ Created on Tue Nov 14 @author: Yee Shuayi This code will preprocess all datasets collected through MTurk """ import os import pandas as pd import numpy as np def remove_non_ascii(text): ''' Function to remove non ascii characters ''' # Remove non ascii characters text = ''.join([i if ord(i) >=32 and ord(i) <=126 else ' ' for i in text]) return text def remove_punctuation(text): # Remove punctuation text = text.translate(str.maketrans('', '', string.punctuation)) return text def remove_numbers(text): # Remove numbers text = ''.join([i for i in text if not i.isdigit()]) return text def convert_to_lowercase(text): # Convert text to lowercase text = text.lower() return text def preprocess_text(text): # Preprocess text text = remove_non_ascii(text) text = remove_punctuation(text) text = remove_numbers(text) text = convert_to_lowercase(text) return text def preprocess_dataset(dataset): # Preprocess dataset dataset['description'] = dataset['description'].apply(preprocess_text) def main(): # os.chdir('C:/Users/Yee Shuayi/Documents/GitHub/Intelligent-System-Project/data/MTurk/') # datasets = os.listdir() # datasets.sort() # for dataset in datasets: # df = pd.read_csv(dataset) # preprocess_dataset(df) # df.to_csv(dataset,index=False) if __name__ == '__main__': # main()<|repo_name|>johnnynorth/Intelligent-System-Project<|file_sep|>/code/preprocess.py # -*- coding: utf-8 -*- """ Created on Tue Nov @author: Yee Shuayi This code will preprocess all datasets collected through MTurk """ import pandas as pd import numpy as np import re from nltk.tokenize import word_tokenize from nltk.stem import WordNetLemmatizer from nltk.corpus import stopwords lemmatizer = WordNetLemmatizer() stop_words = set(stopwords.words('english')) def remove_non_ascii(text): # Remove non ascii characters text = ''.join([i if ord(i) >=32 and ord(i) <=126 else ' ' for i in text]) return text def remove_punctuation(text): # Remove punctuation text = re.sub(r'[^ws]', '',text) return text def remove_numbers(text): # Remove numbers text = ''.join([i for i in text if not i.isdigit()]) return text def convert_to_lowercase(text): # Convert text