**Citation Request: ** Koklu, N., & Sulak, S.A. (2024). Using artificial intelligence techniques for the analysis of obesity status according to the individuals' social and physical activities. Sinop Üniversitesi Fen Bilimleri Dergisi, 9(1), 217-239. https://doi.org/10.33484/sinopfbd.1445215

Obesity Dataset

Obesity is a serious and chronic disease with genetic and environmental interactions. It is defined as an excessive amount of fat tissue in the body that is harmful to health. The main risk factors for obesity include social, psychological, and eating habits. Obesity is a significant health problem for all age groups in the world. Currently, more than 2 billion people worldwide are obese or overweight. Research has shown that obesity can be prevented. In this study, artificial intelligence methods were used to identify individuals at risk of obesity. An online survey was conducted on 1610 individuals to create the obesity dataset. To analyze the survey data, four commonly used artificial intelligence methods in literature, namely Artificial Neural Network, K Nearest Neighbors, Random Forest and Support Vector Machine, were employed after pre-processing. As a result of this analysis, obesity classes were predicted correctly with success rates of 74.96%, 74.03%, 74.03% and 87.82%, respectively. Random Forest was the most successful artificial intelligence method for this dataset and accurately classified obesity with a success rate of 87.82%.

Attributes/Values Sex Male (712) Female (898)

Age Values in integers

Height Values in integers (cm)

Overweight/Obese Families Yes (266) No (1344)

Consumption of Fast Food Yes (436) No (1174)

Frequency of Consuming Vegetables Rarely (400) Sometimes (708) Always (502)

Number of Main Meals Daily 1. 1-2 (444) 3 (928) 3+ (238)

Food Intake Between Meals Rarely (346) Sometimes (564) Usually (417) Always (283)

Smoking Yes (492) No (1118)

Liquid Intake Daily amount smaller than one liter (456) Within the range of 1 to 2 liters (523) In excess of 2 liters (631)

Calculation Of Calorie Intake Yes (286) No (1324)

Physical Exercise No physical activity (206) In the range of 1-2 days (290) In the range of 3-4 days (370) In the range of 5-6 days (358) 6+ days (386)

Schedule Dedicated to Technology Between 0 and 2 hours (382) Between 3 and 5 hours (826) Exceeding five hours (402)

Type of Transportation Used Automobile (660) Motorbike (94) Bike (116) Public transportation (602) Walking (138)

Class Underweight (73) Normal (658) Overweight (592) Obesity (287)

🔍 Dataset Overview:

🧑‍🦱 Age: Age of the individual.

🚻 Gender: Gender of the individual.

📏 Height: Height of the individual.

⚖️ Weight: Weight of the individual.

🏃 CALC: Caloric intake.

🍟 FAVC: Frequent consumption of high-calorie food.

🥗 FCVC: Frequency of vegetable consumption.

🍽️ NCP: Number of main meals.

🍬 SCC: Consumption of sweet drinks.

🚬 SMOKE: Smoking habits.

💧 CH2O: Daily water intake.

🏠 Family history of overweight: Whether there's a family history of overweight.

🏃 FAF: Physical activity frequency.

⌛ TUE: Time using technology devices.

🍕 CAEC: Consumption of food between meals.

🚗 MTRANS: Transportation method.

⚖️ NObeyesdad: Obesity level.

National Obesity Percentages by State. Explanation of Field Attributes:Obesity - The percent of the state population that is considered obese from the 2015 CDC BRFSS Survey.

West Virginia, Mississippi, and Arkansas are the U.S. states with the highest percentage of their population who are obese. The states with the lowest percentage of their population who are obese include Colorado, Hawaii, and Massachusetts. Obesity in the United States Obesity is a growing problem in many countries around the world, but the United States has the highest rate of obesity among all OECD countries. The prevalence of obesity in the United States has risen steadily over the previous two decades, with no signs of declining. Obesity in the U.S. is more common among women than men, and overweight and obesity rates are higher among African Americans than any other race or ethnicity. Causes and health impacts Obesity is most commonly the result of a combination of poor diet, overeating, physical inactivity, and a genetic susceptibility. Obesity is associated with various negative health impacts, including an increased risk of cardiovascular diseases, certain types of cancer, and diabetes type 2. As of 2022, around 8.4 percent of the U.S. population had been diagnosed with diabetes. Diabetes is currently the eighth leading cause of death in the United States.

Contains tabulated outputs for obesity-related hospital admissions and prescriptions for the treatment of obesity.

In the United States, the rate of obesity is lower among college graduates compared to those who did not graduate from college. For example, in 2023, around 27 percent of college graduates were obese, while 36 percent of those with some college or technical school were obese. At that time, rates of obesity were highest among those with less than a high school education, at around 37 percent. Income and obesity As with education level, there are also differences in rates of obesity in the United States based on income. Adults in the U.S. with an annual income of 75,000 U.S. dollars or more have the lowest rates of obesity, with around 29 percent of this population obese in 2023. On the other hand, those earning less than 15,000 U.S. dollars per year had the highest rates of obesity at that time, at 37 percent. One reason for this disparity may be a lack of access to fresh food among those earning less, as cheap food in the United States tends to be unhealthier. What is the most obese state? As of 2023, the states with the highest rates of obesity were West Virginia, Mississippi, and Arkansas. At that time, around 41 percent of adults in West Virginia were obese. The states with the lowest rates of obesity were Colorado, Hawaii, and Massachusetts. Still, around a quarter of adults in Colorado were obese in 2023. West Virginia and Mississippi are also the states with the highest rates of obesity among high school students. Children with obesity are more likely to be obese as adults and are at increased risk of health conditions such as asthma, type 2 diabetes, and sleep apnea.

This dataset contains information about obesity levels based on eating behavior and physical conditions of individuals from Dhaka, Bangladesh. This comprehensive dataset is in tabular format that encompasses 17 attributes and 2182 records, providing insights into factors influencing obesity. This dataset is collected by doing surveys. The attributes of the dataset are as follows:

1. Gender
2. Age
3. Height (m)
4. Weight (kg)
5. Family history of overweight
6. High caloric food consumption
7. Vegetable consumption frequency
8. Daily main meals frequency
9. Between-meal food consumption frequency
10. Smoking
11. Alcohol intake
12. Daily water intake
13. Monitor calories
14. Physical exercise
15. Daily device usage duration
16. Mode of transportation
17. Obesity level

This set of files contains public data used to validate the grocery data. All references to the original sources are provided below.CHILD OBESITYPeriodically, the English National Health Service (NHS) publishes statistics about various aspects of the health and habits of people living in England, including obesity. The NHS National Child Measurement (NCMP) measures the height and weight of children in Reception class (aged 4 to 5) and year 6 (aged 10 to 11), to assess overweight and obesity levels in children within primary schools. The program is carried out every year in England and statistics are produced at the level of Local Authority (that corresponds to Boroughs in London). We report the data for the school year 2015-2016 (file: child_obesity_london_borough_2015-2016.csv). For the school year 2013-2014, statistics in London are also available at ward-level (file: child_obesity_london_ward_2013-2014.csv)The files are comma-separated and contain the following fields: area_id: the id of the boroughnumber_reception_measured: number of children in reception year measurednumber_y6_measured: number of children in reception year measuredprevalence_overweight_reception: the prevalence (percentage) of overweight children in reception year prevalence_overweight_y6: the prevalence (percentage) of overweight children in year 6prevalence_obese_reception: the prevalence (percentage) of obese children in reception yearprevalence_obese_y6: the prevalence (percentage) of obese children in year 6ADULT OBESITYThe Active People Survey (APS) was a survey used to measure the number of adults taking part in sport across England and included two questions about the height and weight of participants. We report the results of the APS for the year 2012. Prevalence of underweight, healthy weight, overweight, and obese people at borough level are provided in the file london_obesity_borough_2012.csv.The file is comma-separated and contains the following fields: area_id: the id of the boroughnumber_measured: number of people who participated in the surveyprevalence_healthy_weight: the prevalence (percentage) of healthy-weight peopleprevalence_overweight: the prevalence (percentage) of overweight peopleprevalence_obese: the prevalence (percentage) of obese peopleBARIATRIC HOSPITALIZATIONThe NHS records and publishes an annual compendium report about the number of hospital admissions attributable to obesity or bariatric surgery (i.e., weight loss surgery used as a treatment for people who are very obese), and the number of prescription items provided in primary care for the treatment of obesity. The NHS provides both raw counts at the Local Authority level and numbers normalized by population living in those areas. In the file obesity_hospitalization_borough_2016.csv, we report the statistics for the year 2015 (measurements made between Jan 2015 and March 2016).The file is comma-separated and contains the following fields:area_id: the id of the boroughtotal_hospitalizations: total number of obesity-related hospitalizationstotal_bariatric: total number of hospitalizations for bariatric surgeryprevalence_hospitalizations: prevalence (percentage) of obesity-related hospitalizations prevalence_bariatric: prevalence (percentage) of bariatric surgery hospitalizations DIABETESThrough the Quality and Outcomes Framework, NHS Digital publishes annually the number of people aged 17+ on a register for diabetes at each GP practice in England. NHS also publishes the number of people living in a census area who are registered to any of the GP in England. Based on these two sources, an estimate is produced about the prevalence of diabetes in each area. The data (file diabetes_estimates_osward_2016.csv) was collected in 2016 at LSOA-level and published at ward-level.The file is comma-separated and contains the following fields:area_id: the id of the wardgp_patients: total number of GP patients gp_patients_diabetes: total number of GP patients with a diabetes diagnosisestimated_diabetes_prevalence: prevalence (percentage) of diabetesAREA MAPPINGMapping of Greater London postcodes into larger geographical aggregations. The file is comma-separated and contains the following fields:pcd: postcodelat: latitudelong: longitudeoa11: output arealsoa11: lower super output areamsoa11: medium super output areaosward: wardoslaua: borough

Obesity

Obesity, which causes physical and mental problems, is a global health problem with serious consequences. The prevalence of obesity is increasing steadily, and therefore, new research is needed that examines the influencing factors of obesity and how to predict the occurrence of the condition according to these factors.

" https://www.semanticscholar.org/paper/Estimation-of-Obesity-Levels-with-a-Trained-Neural-Ya%C4%9F%C4%B1n-G%C3%BCl%C3%BC/2c1eab51db154493d225c8b86ba885bbaf147a2c "

Dataset Information

This dataset include data for the estimation of obesity levels in individuals from the countries of Mexico, Peru and Colombia, based on their eating habits and physical condition. The data contains 17 attributes and 2111 records, the records are labeled with the class variable NObesity (Obesity Level), that allows classification of the data using the values of Insufficient Weight, Normal Weight, Overweight Level I, Overweight Level II, Obesity Type I, Obesity Type II and Obesity Type III. 77% of the data was generated synthetically using the Weka tool and the SMOTE filter, 23% of the data was collected directly from users through a web platform.

Gender: Feature, Categorical, "Gender" Age : Feature, Continuous, "Age"
Height: Feature, Continuous
Weight: Feature Continuous
family_history_with_overweight: Feature, Binary, " Has a family member suffered or suffers from overweight? "

FAVC : Feature, Binary, " Do you eat high caloric food frequently? "
FCVC : Feature, Integer, " Do you usually eat vegetables in your meals? "
NCP : Feature, Continuous, " How many main meals do you have daily? "
CAEC : Feature, Categorical, " Do you eat any food between meals? "
SMOKE : Feature, Binary, " Do you smoke? "
CH2O: Feature, Continuous, " How much water do you drink daily? "
SCC: Feature, Binary, " Do you monitor the calories you eat daily? "
FAF: Feature, Continuous, " How often do you have physical activity? "
TUE : Feature, Integer, " How much time do you use technological devices such as cell phone, videogames, television, computer and others? "

CALC : Feature, Categorical, " How often do you drink alcohol? "
MTRANS : Feature, Categorical, " Which transportation do you usually use? "
NObeyesdad : Target, Categorical, "Obesity level"

The adult obesity rate, or the percentage of the county population (age 18 and older*) that is obese, or has a Body Mass Index (BMI) equal to or greater than 30 [kg/m2], is illustrative of a serious health problem, in Champaign County, statewide, and nationally.

The adult obesity rate data shown here spans from Reporting Years (RY) 2015 to 2024. Champaign County’s adult obesity rate fluctuated during this time, peaking in RY 2022. The adult obesity rates for Champaign County, Illinois, and the United States were all above 30% in RY 2024, but the Champaign County rate was lower than the state and national rates. All counties in Illinois had an adult obesity rate above 30% in RY 2024, but Champaign County's rate is one of the lowest among all Illinois counties.

Obesity is a health problem in and of itself, and is commonly known to exacerbate other health problems. It is included in our set of indicators because it can be easily measured and compared between Champaign County and other areas.

This data was sourced from the University of Wisconsin’s Population Health Institute’s and the Robert Wood Johnson Foundation’s County Health Rankings & Roadmaps. Each year’s County Health Rankings uses data from the most recent previous years that data is available. Therefore, the 2024 County Health Rankings (“Reporting Year” in the table) uses data from 2021 (“Data Year” in the table). The survey methodology changed in Reporting Year 2015 for Data Year 2011, which is why the historical data shown here begins at that time. No data is available for Data Year 2018. The County Health Rankings website notes to use caution if comparing RY 2024 data with prior years.

*The percentage of the county population measured for obesity was age 20 and older through Reporting Year 2021, but starting in Reporting Year 2022 the percentage of the county population measured for obesity was age 18 and older.

Source: University of Wisconsin Population Health Institute. County Health Rankings & Roadmaps 2024. www.countyhealthrankings.org.

This dataset shows the levels of overweight and obese people by country. Data is provided for 2002 and 2010 as a percentage of the total population and is also broken down by sex. Rates of change between 2002 and 2010 are also provided. The data was collated by the World Health Organisation (WHO)(http://www.who.int/gho/ncd/risk_factors/overweight/en/index.html) and was downloaded via the Guardian website (http://www.theguardian.com/news/datablog/interactive/2013/feb/19/obesity-map-of-world-weight). GIS vector data. This dataset was first accessioned in the EDINA ShareGeo Open repository on 2014-01-03 and migrated to Edinburgh DataShare on 2017-02-22.

Obesity rates for each Census Tract in Allegheny County were produced for the study “Developing small-area predictions for smoking and obesity prevalence in the United States." The data is not explicitly based on population surveys or data collection conducted in Allegheny County, but rather estimated using statistical modeling techniques. In this technique, researchers applied the obesity rate of a demographically similar census tract to one in Allegheny County to compute an obesity rate.

Note, August 2011: A number of errors have been identified in Table 7.5 - GHQ 12 score by body mass index (BMI) and gender, 2008 on page 62 of the Statistics on Obesity, Physical Activity and Diet: England, 2011 report. The errors also affect the corresponding table in the accompanying Excel workbook. The commentary in the pdf report is unaffected. Please see the errata note for further information and corrected figures. The NHS IC apologises for any inconvenience this may have caused. Summary: This statistical report presents a range of information on obesity, physical activity and diet, drawn together from a variety of sources. The topics covered include: overweight and obesity prevalence among adults and children physical activity levels among adults and children trends in purchases and consumption of food and drink and energy intake health outcomes of being overweight or obese This report contains seven chapters: Chapter 1: Introduction; this summarises Government plans and targets in this area, as well as providing sources of further information and links to relevant documents. Note, many of these were introduced by the previous government but were relevant at the time the data were collected. Chapters 2 to 6 cover obesity, physical activity and diet providing an overview of the key findings from a number of sources of previously published information, whilst maintaining useful links to each section of the reports. Additional analysis has been undertaken of the Health Survey for England (HSE) to provide more detailed information previously unpublished. Chapter 7: Health Outcomes; presents a range of information about the health outcomes of being obese or overweight which includes information on health risks, hospital admissions and prescription drugs used for treatment of obesity. Figures presented in Chapter 7 have been obtained from a number of sources and presented in a user-friendly format. Most of the data contained in the chapter have been published previously by the NHS Information Centre or the National Audit Office. Previously unpublished figures on obesity-related Finished Hospital Episodes and Finished Consultant Episodes for 2009/10 are presented using data from the NHS Information Centre's Hospital Episode Statistics as well as data from the Prescribing Unit at the NHS Information Centre on prescription items dispensed for treatment of obesity.

Estimation of Obesity Levels Based on Eating Habits and Physical Condition

  Overview

This dataset is designed to estimate obesity levels based on several parameters including eating habits, physical condition, and lifestyle. It includes data from diverse individuals across different demographics, offering insights for research in healthcare, nutrition, and public health.

  Dataset Characteristics

Type: Multivariate Number of Instances: 2111 Number of Attributes:… See the full description on the dataset page: https://huggingface.co/datasets/naabiil/Obesity_Levels_Estimation.

Discover the Obesity Prediction Dataset featuring 17 attributes and 2,111 records to analyze eating habits, lifestyle, and obesity levels.

This statistic displays individuals who are obese (as measured by their body mass index (BMI) by gender in Wales from 2003 to 2015. In 2011, 22 percent of men and 22 percent of women were obese.