diff --git a/README.md b/README.md index 313db9d..9133e7f 100644 --- a/README.md +++ b/README.md @@ -1,8 +1,19 @@ -# Bangkit_Project - -This repo for our Bangkit_Project. - -## Novel Corona Virus 2019 Dataset +# PREDICT USING LONG SHORT TERM MEMORY (LSTM) MODELS +This repo for our Bangkit Academy Projects + +## OBJECTIVES +### PROBLEM FRAMING + The predicted growth of people infected from the number of positive indicated Coronavirus. +### HYPOTHESIS + - How does the age range that affects infected coronavirus? + - How does gender affect the number of people infected with coronavirus? + - How many people will be infected by the coronavirus in the next seven days? +### FORMULATION + Predicting using TensorFlow with Long Short-Term Memory (LSTM) Models +### PRESENTATION + This link for our slide presentation (http://bit.ly/LSTMcovid19) + +## NOVEL CORONAVIRUS 2019 DATASET Source : https://www.kaggle.com/larsupb/covid-19-forecast-germany-with-lgbm-and-keras/data Day level information on covid-19 affected cases @@ -62,3 +73,48 @@ Some insights could be - Changes in number of affected cases over time - Change in cases over time at country level - Latest number of affected cases + +## METHODE +#### Data Preparation + +The process of using domain knowledge to extract features from raw data via data mining techniques. + +* Feature Selection +* Numerical Imputation +* Categorical Column Grouping +* Extracting column “age_range” +* Extracting date +* Data Profiling + +#### Techniques + +Long short-term memory (LSTM) is an artificial recurrent neural network (RNN) architecture used in the field of deep learning. Unlike standard feedforward neural networks, LSTM has feedback connections. It can not only process single data points (such as images), but also entire sequences of data (such as speech or video). + +## RESULTS + +![1](covid19/1.png) + +People who are prone to coronavirus are old people and young people, especially older people from 80-99 years old. Based on this result, we conclude that older people more susceptible to coronavirus than younger people. And the number of Female patients is more than Male Patients. + +____ + +![2](covid19/2.png) + +The reason why people get infected by coronavirus. The most common things are direct contact with another patient and Visit to Daegu. Therefore, people who have not been exposed to coronavirus should keep their distance from coronavirus patients, and are requested for not to visit Daegu for a while. + +_____ + +![3](covid19/3.png) + +Graph of total coronavirus cases. + +Based on the data, coronavirus began to spread by the end of February. Day after day, the case of coronavirus is apparently growing and more people are being affected. + +____ + +![4](covid19/4.png) + +Graph of Coronavirus prediction in the next 7 days + +Based on the results we have made, the model predicted that for the next seven days the coronavirus will increase, even exceeding the ten thousand people affected by coronavirus. Based on this result, people should be more careful and take care of their health. + diff --git a/covid19/1.png b/covid19/1.png new file mode 100644 index 0000000..ead03ee Binary files /dev/null and b/covid19/1.png differ diff --git a/covid19/2.png b/covid19/2.png new file mode 100644 index 0000000..8b22416 Binary files /dev/null and b/covid19/2.png differ diff --git a/covid19/3.png b/covid19/3.png new file mode 100644 index 0000000..79d27d8 Binary files /dev/null and b/covid19/3.png differ diff --git a/covid19/4.png b/covid19/4.png new file mode 100644 index 0000000..ba920d6 Binary files /dev/null and b/covid19/4.png differ