Context

Released Under: National Data Sharing and Accessibility Policy (NDSAP)

Content

The catalog contains the details of natural & un-natural accidents, traffic accidents and suicides. Major causes of accidental deaths are Traffic accidents, Drowning, Lightning, Heart Attacks, Accidental Fire, Falls, Poisoning. Major causes of Suicides are Family Problems, Illness, Bankruptcy or Indebtedness, with Percentage Variation Age and Gender-wise Distribution, It includes Data of All India Level, The information given in this report is police recorded data which has been obtained through State/UT Police, It includes Total number of Accidental Deaths with Mid-Year Projected Population and Rate of Accidental Deaths.

For more details about each file, refer the following links.

Source: https://data.gov.in/catalog/accidental-deaths-suicides-india-adsi-2019 Open Government Data (OGD) Platform India

Acknowledgements

Ministry of Home Affairs Department of States National Crime Records Bureau (NCRB)

Vital Statistics: Death Rate: per 1000 Population: Andhra Pradesh: Urban data was reported at 4.900 NA in 2020. This records an increase from the previous number of 4.800 NA for 2019. Vital Statistics: Death Rate: per 1000 Population: Andhra Pradesh: Urban data is updated yearly, averaging 5.400 NA from Dec 1997 (Median) to 2020, with 23 observations. The data reached an all-time high of 6.100 NA in 1998 and a record low of 4.800 NA in 2019. Vital Statistics: Death Rate: per 1000 Population: Andhra Pradesh: Urban data remains active status in CEIC and is reported by Office of the Registrar General & Census Commissioner, India. The data is categorized under India Premium Database’s Demographic – Table IN.GAH003: Vital Statistics: Death Rate: by States.

Number and percentage of deaths, by month and place of residence, 1991 to most recent year.

Context

From World Health Organization - On 31 December 2019, WHO was alerted to several cases of pneumonia in Wuhan City, Hubei Province of China. The virus did not match any other known virus. This raised concern because when a virus is new, we do not know how it affects people.

So daily level information on the affected people can give some interesting insights when it is made available to the broader data science community.

Johns Hopkins University has made an excellent dashboard using the affected cases data. Data is extracted from the google sheets associated and made available here.

Edited: Now data is available as csv files in the Johns Hopkins Github repository. Please refer to the github repository for the Terms of Use details. Uploading it here for using it in Kaggle kernels and getting insights from the broader DS community.

Content

2019 Novel Coronavirus (2019-nCoV) is a virus (more specifically, a coronavirus) identified as the cause of an outbreak of respiratory illness first detected in Wuhan, China. Early on, many of the patients in the outbreak in Wuhan, China reportedly had some link to a large seafood and animal market, suggesting animal-to-person spread. However, a growing number of patients reportedly have not had exposure to animal markets, indicating person-to-person spread is occurring. At this time, it’s unclear how easily or sustainably this virus is spreading between people - CDC

This dataset has daily level information on the number of affected cases, deaths and recovery from 2019 novel coronavirus. Please note that this is a time series data and so the number of cases on any given day is the cumulative number.

The data is available from 22 Jan, 2020.

Column Description

Province/State - Province or state of the observation (Could be empty when missing) CountryReg - Country of observation Last Update - Time in UTC at which the row is updated for the given province or country. (Not standardised and so please clean before using it) Confirmed - Cumulative number of confirmed cases till that date Deaths - Cumulative number of of deaths till that date Recovered - Cumulative number of recovered cases till that date Lon Lat week - Week Number (1 To 52) Weeks Per Year

Acknowledgements

Johns Hopkins University for making the data available for educational and academic research purposes.

Inspiration

Your data will be in front of the world's largest data science community. What questions do you want to see answered?

Coronavirus disease 2019 (COVID-19) time series listing confirmed cases, reported deaths and reported recoveries. Data is disaggregated by country and States. This dataset includes time series data tracking the number of people affected by COVID-19 India Wide, including: - confirmed tested cases of Coronavirus infection the number of people who have reportedly died while sick with Coronavirus the number of people who have reportedly recovered from it.

Accident-Detection-Model

Accident Detection Model is made using YOLOv8, Google Collab, Python, Roboflow, Deep Learning, OpenCV, Machine Learning, Artificial Intelligence. It can detect an accident on any accident by live camera, image or video provided. This model is trained on a dataset of 3200+ images, These images were annotated on roboflow.

Problem Statement

• Road accidents are a major problem in India, with thousands of people losing their lives and many more suffering serious injuries every year.
• According to the Ministry of Road Transport and Highways, India witnessed around 4.5 lakh road accidents in 2019, which resulted in the deaths of more than 1.5 lakh people.
• The age range that is most severely hit by road accidents is 18 to 45 years old, which accounts for almost 67 percent of all accidental deaths.

Accidents survey

https://user-images.githubusercontent.com/78155393/233774342-287492bb-26c1-4acf-bc2c-9462e97a03ca.png" alt="Survey">

Literature Survey

• Sreyan Ghosh in Mar-2019, The goal is to develop a system using deep learning convolutional neural network that has been trained to identify video frames as accident or non-accident.
• Deeksha Gour Sep-2019, uses computer vision technology, neural networks, deep learning, and various approaches and algorithms to detect objects.

Research Gap

• Lack of real-world data - We trained model for more then 3200 images.
• Large interpretability time and space needed - Using google collab to reduce interpretability time and space required.
• Outdated Versions of previous works - We aer using Latest version of Yolo v8.

Proposed methodology

• We are using Yolov8 to train our custom dataset which has been 3200+ images, collected from different platforms.
• This model after training with 25 iterations and is ready to detect an accident with a significant probability.

Model Set-up

Preparing Custom dataset

• We have collected 1200+ images from different sources like YouTube, Google images, Kaggle.com etc.
• Then we annotated all of them individually on a tool called roboflow.
• During Annotation we marked the images with no accident as NULL and we drew a box on the site of accident on the images having an accident
• Then we divided the data set into train, val, test in the ratio of 8:1:1
• At the final step we downloaded the dataset in yolov8 format.
  #### Using Google Collab
• We are using google colaboratory to code this model because google collab uses gpu which is faster than local environments.
• You can use Jupyter notebooks, which let you blend code, text, and visualisations in a single document, to write and run Python code using Google Colab.
• Users can run individual code cells in Jupyter Notebooks and quickly view the results, which is helpful for experimenting and debugging. Additionally, they enable the development of visualisations that make use of well-known frameworks like Matplotlib, Seaborn, and Plotly.
• In Google collab, First of all we Changed runtime from TPU to GPU.
• We cross checked it by running command ‘!nvidia-smi’
  #### Coding
• First of all, We installed Yolov8 by the command ‘!pip install ultralytics==8.0.20’
• Further we checked about Yolov8 by the command ‘from ultralytics import YOLO from IPython.display import display, Image’
• Then we connected and mounted our google drive account by the code ‘from google.colab import drive drive.mount('/content/drive')’
• Then we ran our main command to run the training process ‘%cd /content/drive/MyDrive/Accident Detection model !yolo task=detect mode=train model=yolov8s.pt data= data.yaml epochs=1 imgsz=640 plots=True’
• After the training we ran command to test and validate our model ‘!yolo task=detect mode=val model=runs/detect/train/weights/best.pt data=data.yaml’ ‘!yolo task=detect mode=predict model=runs/detect/train/weights/best.pt conf=0.25 source=data/test/images’
• Further to get result from any video or image we ran this command ‘!yolo task=detect mode=predict model=runs/detect/train/weights/best.pt source="/content/drive/MyDrive/Accident-Detection-model/data/testing1.jpg/mp4"’
• The results are stored in the runs/detect/predict folder.
  Hence our model is trained, validated and tested to be able to detect accidents on any video or image.

Challenges I ran into

I majorly ran into 3 problems while making this model

• I got difficulty while saving the results in a folder, as yolov8 is latest version so it is still underdevelopment. so i then read some blogs, referred to stackoverflow then i got to know that we need to writ an extra command in new v8 that ''save=true'' This made me save my results in a folder.
• I was facing problem on cvat website because i was not sure what

This data contains all the essential data in the form of % with respect to rural and urban Indian states . This dataset is highly accurate as this is taken from the Indian govt. it is updated till 2021 for all states and union territories. source of data is data.gov.in titled - ******All India and State/UT-wise Factsheets of National Family Health Survey******

it is advised to you pls search the data keywords you need by using (Ctrl+f) , as it will help to avoid time wastage. States/UTs

Different columns it contains are Area

Number of Households surveyed Number of Women age 15-49 years interviewed Number of Men age 15-54 years interviewed

Female population age 6 years and above who ever attended school (%)

Population below age 15 years (%)

Sex ratio of the total population (females per 1,000 males)

Sex ratio at birth for children born in the last five years (females per 1,000 males)

Children under age 5 years whose birth was registered with the civil authority (%)

Deaths in the last 3 years registered with the civil authority (%)

Population living in households with electricity (%)

Population living in households with an improved drinking-water source1 (%)

Population living in households that use an improved sanitation facility2 (%)

Households using clean fuel for cooking3 (%) Households using iodized salt (%)

Households with any usual member covered under a health insurance/financing scheme (%)

Children age 5 years who attended pre-primary school during the school year 2019-20 (%)

Women (age 15-49) who are literate4 (%)

Men (age 15-49) who are literate4 (%)

Women (age 15-49) with 10 or more years of schooling (%)

Men (age 15-49) with 10 or more years of schooling (%)

Women (age 15-49) who have ever used the internet (%)

Men (age 15-49) who have ever used the internet (%)

Women age 20-24 years married before age 18 years (%)

Men age 25-29 years married before age 21 years (%)

Total Fertility Rate (number of children per woman) Women age 15-19 years who were already mothers or pregnant at the time of the survey (%)

Adolescent fertility rate for women age 15-19 years5 Neonatal mortality rate (per 1000 live births)

Infant mortality rate (per 1000 live births) Under-five mortality rate (per 1000 live births)

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - Any method6 (%)

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - Any modern method6 (%)

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - Female sterilization (%)

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - Male sterilization (%)

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - IUD/PPIUD (%)

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - Pill (%)  

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - Condom (%)

Current Use of Family Planning Methods (Currently Married Women Age 15-49 years) - Injectables (%)

Total Unmet need for Family Planning (Currently Married Women Age 15-49 years)7 (%)

Unmet need for spacing (Currently Married Women Age 15-49 years)7 (%)

Health worker ever talked to female non-users about family planning (%)

Current users ever told about side effects of current method of family planning8 (%)

Mothers who had an antenatal check-up in the first trimester (for last birth in the 5 years before the survey) (%)

Mothers who had at least 4 antenatal care visits (for last birth in the 5 years before the survey) (%)

Mothers whose last birth was protected against neonatal tetanus (for last birth in the 5 years before the survey)9 (%)

Mothers who consumed iron folic acid for 100 days or more when they were pregnant (for last birth in the 5 years before the survey) (%)

Mothers who consumed iron folic acid for 180 days or more when they were pregnant (for last birth in the 5 years before the survey} (%)

Registered pregnancies for which the mother received a Mother and Child Protection (MCP) card (for last birth in the 5 years before the survey) (%)

Mothers who received postnatal care from a doctor/nurse/LHV/ANM/midwife/other health personnel within 2 days of delivery (for last birth in the 5 years before the survey) (%)

Average out-of-pocket expenditure per delivery in a public health facility (for last birth in the 5 years before the survey) (Rs.)

Children born at home who were taken to a health facility for a check-up within 24 hours of birth (for last birth in the 5 years before the survey} (%)

Children who received postnatal care from a doctor/nurse/LHV/ANM/midwife/ other health personnel within 2 days of delivery (for last birth in the 5 years before the survey) (%)

Institutional births (in the 5...

Context

Covid-19 Data collected from various sources on the internet. This dataset has daily level information on the number of affected cases, deaths, and recovery from the 2019 novel coronavirus. Please note that this is time-series data and so the number of cases on any given day is the cumulative number.

Content

The dataset includes 28 files scrapped from various data sources mainly the John Hopkins GitHub repository, the ministry of health affairs India, worldometer, and Our World in Data website. The details of the files are as follows

• countries-aggregated.csv A simple and cleaned data with 5 columns with self-explanatory names. -covid-19-daily-tests-vs-daily-new-confirmed-cases-per-million.csv A time-series data of daily test conducted v/s daily new confirmed case per million. Entity column represents Country name while code represents ISO code of the country. -covid-contact-tracing.csv Data depicting government policies adopted in case of contact tracing. 0 -> No tracing, 1-> limited tracing, 2-> Comprehensive tracing. -covid-stringency-index.csv The nine metrics used to calculate the Stringency Index are school closures; workplace closures; cancellation of public events; restrictions on public gatherings; closures of public transport; stay-at-home requirements; public information campaigns; restrictions on internal movements; and international travel controls. The index on any given day is calculated as the mean score of the nine metrics, each taking a value between 0 and 100. A higher score indicates a stricter response (i.e. 100 = strictest response). -covid-vaccination-doses-per-capita.csv A total number of vaccination doses administered per 100 people in the total population. This is counted as a single dose, and may not equal the total number of people vaccinated, depending on the specific dose regime (e.g. people receive multiple doses). -covid-vaccine-willingness-and-people-vaccinated-by-country.csv Survey who have not received a COVID vaccine and who are willing vs. unwilling vs. uncertain if they would get a vaccine this week if it was available to them. -covid_india.csv India specific data containing the total number of active cases, recovered and deaths statewide. -cumulative-deaths-and-cases-covid-19.csv A cumulative data containing death and daily confirmed cases in the world. -current-covid-patients-hospital.csv Time series data containing a count of covid patients hospitalized in a country -daily-tests-per-thousand-people-smoothed-7-day.csv Daily test conducted per 1000 people in a running week average. -face-covering-policies-covid.csv Countries are grouped into five categories: 1->No policy 2->Recommended 3->Required in some specified shared/public spaces outside the home with other people present, or some situations when social distancing not possible 4->Required in all shared/public spaces outside the home with other people present or all situations when social distancing not possible 5->Required outside the home at all times regardless of location or presence of other people -full-list-cumulative-total-tests-per-thousand-map.csv Full list of total tests conducted per 1000 people. -income-support-covid.csv Income support captures if the government is covering the salaries or providing direct cash payments, universal basic income, or similar, of people who lose their jobs or cannot work. 0->No income support, 1->covers less than 50% of lost salary, 2-> covers more than 50% of the lost salary. -internal-movement-covid.csv Showing government policies in restricting internal movements. Ranges from 0 to 2 where 2 represents the strictest. -international-travel-covid.csv Showing government policies in restricting international movements. Ranges from 0 to 2 where 2 represents the strictest. -people-fully-vaccinated-covid.csv Contains the count of fully vaccinated people in different countries. -people-vaccinated-covid.csv Contains the total count of vaccinated people in different countries. -positive-rate-daily-smoothed.csv Contains the positivity rate of various countries in a week running average. -public-gathering-rules-covid.csv Restrictions are given based on the size of public gatherings as follows: 0->No restrictions 1 ->Restrictions on very large gatherings (the limit is above 1000 people) 2 -> gatherings between 100-1000 people 3 -> gatherings between 10-100 people 4 -> gatherings of less than 10 people -school-closures-covid.csv School closure during Covid. -share-people-fully-vaccinated-covid.csv Share of people that are fully vaccinated. -stay-at-home-covid.csv Countries are grouped into four categories: 0->No measures 1->Recommended not to leave the house 2->Required to not leave the house with exceptions for daily exercise, grocery shopping, and ‘essent...

Context

What Is COVID-19?

A coronavirus is a kind of common virus that causes an infection in your nose, sinuses, or upper throat. Most coronaviruses aren't dangerous.

COVID-19 is a disease that can cause what doctors call a respiratory tract infection. It can affect your upper respiratory tract (sinuses, nose, and throat) or lower respiratory tract (windpipe and lungs). It's caused by a coronavirus named SARS-CoV-2.

It spreads the same way other coronaviruses do, mainly through person-to-person contact. Infections range from mild to serious.

SARS-CoV-2 is one of seven types of coronavirus, including the ones that cause severe diseases like Middle East respiratory syndrome (MERS) and sudden acute respiratory syndrome (SARS). The other coronaviruses cause most of the colds that affect us during the year but aren’t a serious threat for otherwise healthy people.

In early 2020, after a December 2019 outbreak in China, the World Health Organization identified SARS-CoV-2 as a new type of coronavirus. The outbreak quickly spread around the world.

Is there more than one strain of SARS-CoV-2?

It’s normal for a virus to change, or mutate, as it infects people. A Chinese study of 103 COVID-19 cases suggests the virus that causes it has done just that. They found two strains, which they named L and S. The S type is older, but the L type was more common in early stages of the outbreak. They think one may cause more cases of the disease than the other, but they’re still working on what it all means.

How long will the coronavirus last?

It’s too soon to tell how long the pandemic will continue. It depends on many things, including researchers’ work to learn more about the virus, their search for a treatment and a vaccine, and the public’s efforts to slow the spread.

Dozens of vaccine candidates are in various stages of development and testing. This process usually takes years. Researchers are speeding it up as much as they can, but it still might take 12 to 18 months to find a vaccine that works and is safe.

Symptoms of COVID-19

The main symptoms include:

• Fever
• Coughing
• Shortness of breath
• Fatigue
• Chills, sometimes with shaking
• Body aches
• Headache
• Sore throat
• Loss of smell or taste
• Nausea
• Diarrhea

The virus can lead to pneumonia, respiratory failure, septic shock, and death. Many COVID-19 complications may be caused by a condition known as cytokine release syndrome or a cytokine storm. This is when an infection triggers your immune system to flood your bloodstream with inflammatory proteins called cytokines. They can kill tissue and damage your organs.

STAY HOME. STAY SAFE !

Content

ALL DATASETS HAVE BEEN CLEANED FOR DIRECT USE.

Total_World_covid-19.csv : This dataset contains the worldwide data country-wise such as total cases , total active, deaths, etc. along with testing data.

Total_India_covid-19.csv : This dataset contains India level data statewise such as confirmed cases , active cases, deaths, etc.

Total_US_covid-19.csv : This dataset contains India level data statewise such as confirmed cases , active cases, deaths, etc.

Daily_States_India.csv : This dataset contains daily statewise data of India such as daily confirmed , daily active , daily deaths and daily recovered.

Total_Maharshtra_covid-19.csv : This dataset contains Maharashtra's district wise data such as confirmed cases , active cases, deaths, etc.

Acknowledgements

1. World and US data has been collected from Worldometer . Thanks a lot.

2. India and State level along with Maharashtra district data has been collected from Covid19India. Special thanks to them for providing updated and such wonderful data .

Inspiration

1) What has been the Covid-19 trend across the world, Is it declining? Is it increasing? 2) Which countries have been able to sustain and control the virus spread? 3) How is India coping up with the virus? Have they been able to control it at the given cost of 2 months nationwide lockdown?

“India tops the world with 11% of global death in road accidents It has 1 per cent of the world's vehicles but accounts for 11 per cent of all road crash deaths, witnessing 53 road crashes every hour; killing 1 person every 4 minutes” – World Bank Report -The Economic Times, Feb 14, 2021

This is indeed an alarming report. With the rapid increase in the number of automobiles and also the heavily congested roads, ensuring road safety has utmost importance for the people in the country. Fatalities and injuries resulting from road traffic accidents can create a heavy burden on the economy and can strain the health, insurance and legal systems of the country.

India has recorded a significant drop in road crashes and deaths in in 2020 when compared to 2019. This is partly due to COVID-19 pandemic-prompted lockdowns and people voluntarily not venturing out to prevent transmission. Hence to get a better view, a detailed analysis was carried out to understand some of the major reasons for the high accident cases in India using the data from 2017-2019. The source of this data is from the official website of Government of India (https://data.gov.in/ )

Coronavirus disease 2019 (COVID-19)

Coronavirus disease 2019 (COVID-19) is an infectious disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). The disease was first identified in 2019 in Wuhan, China, and has since spread globally, resulting in the 2019–20 coronavirus pandemic. Epidemiologists are teaming up with data scientists to stem the spread of the novel coronavirus by tapping big data, machine learning and other digital tools. The goal is to get real-time forecasts and other critical information to front-line health-care workers and public policy makers as the outbreak unfolds. The objective of the Hackathon is to predict the probability of person getting infected by Covid-19.

Coronaviruses are a family of hundreds of viruses that can cause fever, respiratory problems, and sometimes gastrointestinal symptoms too. The 2019 novel coronavirus is one of seven members of this family known to infect humans, and the third in the past three decades to jump from animals to humans. Since emerging in China in December, this new coronavirus has caused a global health emergency, sickening almost 200,000 people worldwide, and so far killing more than 9,000. As of March 19, about 10000 cases had been reported in the US, and 155 people have died. In Wuhan, home to 11 million people, the initial number of cases was 40, estimated by a group of researchers led by Natsuko Imai of Imperial College. The number of exposed was assumed to be 20 times this number. The basic reproduction number (BRN) is the expected number of cases directly generated by one case. A BRN greater than one indicates that the outbreak is self-sustaining, while a BRN less than one indicates that the number of new cases decreases over time and eventually the outbreak will stop. Ideally, the BRN should be reduced in order to slow down an epidemic. The BRN in the first three phases was estimated to be 3.1, 2.6, and 1.9, respectively. In the Cell Discovery article, the BRN is assumed to have decreased to 0.9 or 0.5 in phase IV, based on previous experience in SARS. According to an article in Science in 2003, the BRN of SARS decreased from 2.7 to 0.25 after the patients were isolated and the infection started being controlled. The better we can track the virus, the better we can fight it. By analyzing different parameters responsible for the outbreak of coronavirus, we can take controlling measures in an accelerated way.

Part -01 :

The objective of the first part of the problem statement is to predict the probability of a person getting infected by Covid-19 on 20 th March 2020. The output file 01 should contain only people_ID and the respective infect_prob for the test data.

Part -02 :

The Diuresis of a person is a time-dependent parameter, for which you have to come up with a Time-series prediction model. Using the Diuresis predicted by the model, you need to calculate the infect_prob on 27 th March 2020 for every people_ID in the test data. . The output file 02 should contain only people_ID and the respective infect_prob on 27 th March.

Vital Statistics: Death Rate: per 1000 Population: Rajasthan: Rural data was reported at 5.800 NA in 2020. This records a decrease from the previous number of 6.000 NA for 2019. Vital Statistics: Death Rate: per 1000 Population: Rajasthan: Rural data is updated yearly, averaging 7.000 NA from Dec 1997 (Median) to 2020, with 23 observations. The data reached an all-time high of 9.300 NA in 1998 and a record low of 5.800 NA in 2020. Vital Statistics: Death Rate: per 1000 Population: Rajasthan: Rural data remains active status in CEIC and is reported by Office of the Registrar General & Census Commissioner, India. The data is categorized under India Premium Database’s Demographic – Table IN.GAH003: Vital Statistics: Death Rate: by States.