The rapid spread of cardiovascular disease among the world's population is often associated with high-calorie diets, which humans prefer. Such a systemic overload of the human body with high-calorie foods leads to problems with excess weight [1][2][3][4]. In the face of rapid globalization challenges related to climate change, the sustainable development of human capital, according to well-known scientists [5][6][7][8][9], has encountered significant obstacles to the growth of cardiovascular disease [10][11][12][13]. Each person needs a special diet that would consider the body's needs for a certain amount of calories [14][15][16][17][18][19]. The urgency of this problem is growing every year, as evidenced by data from the World Health Organization, both on the growing number of overweight people and the highest mortality worldwide due to cardiovascular disease [3,4]. An unresolved problem in choosing the proper diet to improve the health of any person is establishing data on the caloric content of various foods. Comparative analysis of the caloric content of food will allow patients with cardiovascular disease to properly assess their diet and choose a diet that meets the characteristics of the disease, body condition, and doctors' recommendations.
Objective. To conduct a comparative analysis of the caloric content of food, which will allow people who suffer from cardiovascular disease or are overweight to evaluate their diet and choose the proper diet to improve their health, especially with cardiovascular diseases; promote greater availability of data on the caloric content of foods; to improve the effectiveness of their perception by stakeholders; to establish the relationship between the category of the product and its caloric content; propose an approach to the formation of a diet based on the caloric content of food for people with cardiovascular disease.
Problem proper food selection exists for many other diseases, such as congenital or acquired allergies, cancer, diabetes, etc. [11,14,15,[20][21][22][23][24][25][26][27][28]. With the help of the data obtained from our research, people will choose the foods and calories that will promote a healthy diet. Comparative analysis of the caloric content of food will allow people to find an alternative to those products that cause them an allergic reaction, disrupt the proper functioning of the gastrointestinal tract, and more. The problem of proper nutrition is directly related to the quality of food and its caloric content.
To properly organize diets, you need to understand the essence of caloric foods and their role in human life. Caloric content, as indicated in [29], is an essential indicator of the nutritional value of foods. It is measured in kilocalories (kcal) or kilojoules (kJ). Often, the cause of overweight problems is that people do not understand the caloric content of their foods. The amount of food needed cannot be determined by weight, volume, or appetite because they do not characterize its nutritional value. The only correct criterion for the quantitative side of nutrition is its energy value. It is measured, like energy expenditure, in kilocalories (kcal).
The dataset choice [30][31][32][33][34][35] was due to data on the number of calories per 100 grams of a particular food product. A comparative analysis of the caloric food content is conducted based on a dataset on the caloric content of the products themselves and the quality requirements of conventional products. This study will help deepen people's understanding of the processes through which their body receives energy from the consumption of daily meals and a variety of snacks. Paying particular attention to the number of calories in the food they ate. Readers will have the opportunity to choose a diet that can improve their health.
Consider the attributes of this dataset [30]:
• FoodCategory -this contains data on the category of the food product;
• FoodItem -food product;
• per100grams -the amount of food;
• Cals_per100grams -the number of calories per 100 grams of food;
• KJ_per100grams -the number of kilojoules per 100 grams of food.
As noted by P. Menzel and F. D'Aluisio in [14], people's diets are shaped by their mentality, traditions, and wealth. To confirm their hypothesis, the researchers [14] travelled to about 80 countries to study the national characteristics of each country's diet and identify factors that affect people's diet. After surveying a significant number of respondents, researchers [14] calculated calories per day for the inhabitants of each country and assessed the nutritional value of the food they eat. The scale of the study is impressive not only in the number of people interviewed but also in the duration of it -about three years.
As an example, we can cite several results [14]. A seamstress from Bangladesh has a relatively modest daily diet with a caloric content of 1800 kcal, slightly lower than a woman's norm [14]. Marble Moahi, 32, from poor Botswana, eats very poorly. Her calorie intake per day is only 900, almost three times less than the norm for a woman her age. One look at these questionable foods that she eats during the day is enough to understand that a woman lacks vitamins and nutrients essential for the body's normal functioning. Such limited nutrition is a global problem for third-world countries, affecting their life level and duration [14]. But 54-year-old American Conrad Tolba is moving. The energy value of his diet, which consists mainly of unhealthy fast food, is 5400 kcal! Given that Mr Tolba works as a truck driver and does not lead a very active lifestyle, it is possible to suspect the presence of obesity and problems with the endocrine, cardiovascular and digestive systems [14].
Canadian Eskimo Willy Ishulutak, a 29-year-old professional stone carver [14], is also high in calories. His daily menu with 4,700 calories includes porridge, protein-rich meat, fish, dairy products, and simple carbohydrates in the form of cookies. Given that the weight of a Canadian is average, 64 kg at the height of 1 m 75 cm, this need for extra calories can be explained by a demanding profession [14]. Saleh Abdul is a 32-year-old camel seller from Egypt [14]. He eats a reasonably balanced diet. His menu is not a day, the energy value of which is 3,200 kcal, includes cereals and cereals, flour cakes, vegetables, meat dishes, and chicken eggs. Saleh's life takes place outdoors, so he is allowed to consume energy a little over the norm for a man his age [14]. Leland Melvin is an astronaut [14]. In his diet -not what he wants and loves. Leland's daily menu is carefully designed to fully meet the energy, nutrient, and nutrient needs of a 45-year-old man. The energy value of "space" food is 2700 kcal [14].
To solve the problems in this work, we use standard methods, including the following [36][37][38][39][40][41][42][43]. 1. The correlation field is a graphical representation of statistical indicators, which gives an idea of the relationship between the studied features. We construct a correlation field, where the values of the factor characteristic (x) are plotted on the horizontal axis, and the resultant characteristic (y) is plotted on the vertical axis. They were postponing the correlation field at the intersection of the corresponding values of x and y. The nature of the location of points on the correlation field concludes the direction and form of communication [44][45][46][47][48][49]. If the points are randomly placed throughout the field, there is no relationship between the studied features. The nature of the distribution of points on the correlation field indicates the presence, shape, and direction of the relationship between factor and result characteristics. With the help of the graph, a conclusion is made about the choice of the type of mathematical equation for quantitative estimation of the connection (construction of a regression model by function (regression equation)) [44][45][46][47][48][49].
2. Correlation coefficient -used to estimate the degree of linear dependence between two variables x and y. We can take values from -1 to +1. With positive values of the correlation coefficient with increasing factor variable x increases the resulting variable y; with negative -with increasing factor variable x, the average value of the resulting variable decreases y [43][44][45][46][47][48][49].
3. The correlation ratio reflects the relative degree of variation (variance) of group means (y). The correlation ratio always has a positive value and ranges from 0 to + 1. It takes zero value when the relationship between the studied features is absent [42][43][50][51][52][53][54].
4. Autocorrelation is the correlation of a function with itself shifted by a certain amount of independent variable. Autocorrelation is used to find patterns in several data, such as periodicity [42][43][50][51][52][53][54][55][56][57][58]. It is often used in statistics and signal processing to analyze functions or data series. We could obtain the autocorrelation function graph by plotting the correlation coefficient of two functions (functions shifted by the value and basic one) on the ordinate axis y [55]. If the original function is strictly periodic, then the graph of the autocorrelation function will also be strictly periodic. Thus, from this graph, we can judge the periodicity of the essential function and hence its frequency characteristics [40][41][42][43][56][57][58].
5. A correlation matrix is a matrix of correlation coefficients of several random variables [41,47,53].
6. The multiple correlation coefficient (R) characterizes the closeness of the relationship between the performance indicator and the set of factor indicators [53][54][55][56][57][58].
Data generation was in the form of Excel. Dataset [30] consists of data on the caloric content of food. The following five attributes of the calories in the food items (per 100 grams) data set [30] are presented in Fig. 1. The Calories in Food Items (per 100 grams) [30] consist of the Calories/KJ for 2225 food items per 100 grams across 44 food categories.
The average (arithmetic mean) is one of the main characteristics of the sample, which is the value of the characteristic, the sum of deviations from which all sample values are zero (including deviations). In other words, this is some average level of all measurements for a given group of numbers. It is determined by the formula [41,59].
(
where n is the sample size. We calculated the average values for each of the 44 product categories and presented them in the Cartesian coordinate system (Fig. 2) and the polar coordinate system (Fig. 3). Mode, denoted by Mo, is one of the numerical characteristics of the probability distribution of a random variable. This value occurs in this series more often than in others [41,59].
Median, denoted by Me, is in the mean of the mean when a series of numbers, ordered in ascending or descending order, take the average number (if the series consists of an odd number of numbers) or the arithmetic means of two averages (if the series consists from an even number of numbers) [41,59].
Interval is an indicator that indicates the width of the range of values. It is the difference between the largest and smallest of these numbers.
The sampling variance is used to vary the indicator values relative to the average [41,59].
(
The coefficient of variation [59][60][61] is a relative value that characterizes the scattering (variability) of the trait. Represents the ratio of the standard deviation of to the arithmetic mean, expressed as a percentage.
Skewness [59][60][61] is an indicator that reflects the skew of the distribution relative to the fashion to the left or right. It is the case when some of the reasons contribute to the more frequent occurrence of values that are greater or, conversely, less than the arithmetic mean. Lower values are more common for left-handed or positive asymmetry in the distribution than right-handed or negative.
Kurtosis [59][60][61] is an indicator that reflects the height of the distribution. When any reasons promote the emergence of close to average values, the distribution with positive excess is formed. Suppose extreme values dominate the distribution and, at the same time, lower and higher. In that case, such a distribution is characterized by harmful excess, and in the center of the distribution may form a depression, which turns it into two vertices.
The descriptive statistics analysis tool creates a one-dimensional statistical report containing information about the initial range data's central trend and variability (Table 1).
A histogram [59][60][61] is a diagram consisting of rectangles without gaps between them. Quantitative ratios of some indicators are presented in the form of rectangles, the areas of which are proportional. Most often, for ease of perception, the width of the rectangles is the same, and their height determines the ratio of the displayed parameter. Fig. 4 shows the histogram of the distribution of average calories.
One way to study the distribution series is to construct a cumulative. It allows you to graph the dependence of the value of the feature on the accumulated frequency. The cumulative, or polygon of accumulated frequencies, is often used to represent discrete data. To build a cumulative in Excel, enter data in rows or columns, click "insert" -"chart." Select one of the appropriate scatter charts, indicates the required data for construction (only two lines -the value of the sign and the accumulated frequency), and click "done" [59][60][61]. Fig. 5 shows cumulative frequency of average calories.
The moving average method is the simplest way to smooth empirical curves [59][60][61]. The essence of this method is to replace the actual values of the indicator with their average values, which have much less variation than the initial levels of the series [59][60][61]. Simple moving averages are calculated for odd and even numbers of time intervals depending on the averaging period. Consider the procedure for constructing a moving average with an odd number of members.
A time series consists of levels y1, y2, y3,…, yn.
To determine the moving average, calculate the sum of m elements of the series (where m is an odd number), gradually moving from the first terms y1, y2, y3,…, yn to the following groups of levels: y 2, y 3,… y m + 1; y 3, y 4,… y m + 2; y 4, y 5,… y m + 3; etc.
The arithmetic means are determined for individual sums, changing their value ("slides") as the parameter t increases. A new time series is formed from arithmetic averages, the elements of which are primarily free from random external influences on the forecast indicator. It is believed that moving averages more accurately characterize the tendency to change the feature than the levels of the original time series. Three-and five-membered averages are most often used in practice [59][60][61].
When constructing a weighted moving average on each active section, the value of the main level is replaced by the calculated one, determined by the formula of the weighted arithmetic mean [59][60][61]. In other words, the weighted variable average differs from the simple moving average in that the levels included in the averaging interval are summed with different weights.
A simple moving average considers all levels of the series included in the smoothing interval with equal weights. The weighted average assigns to each level a weight that depends on the distance of this level to the level in the middle of the interval. For a simple moving average in the smoothing interval, the calculation is performed based on a straight line -a first-order polynomial-and smoothing a weighted moving average using polynomials of higher orders, preferably second or second third. Therefore, the simple moving average method can be considered a particular case of the weighted moving average [59][60][61]. Fig. 6 -Fig. 12 show the linear smoothing at w = 3, 5, 7, 9, 11, 13, 15 according. Properties of moving average 1. When using the method of moving averages, choosing the value of the smoothing interval should be made based on substantive considerations and tied to the period of possibly existing oscillatory processes. If the average moving procedure is used to smooth the time series in the absence of any fluctuations, the value of the smoothing interval is often chosen equal to three, five, or seven. The larger the averaging interval (smoothing), the smoother the trend chart looks [59][60][61].
2. Neighboring members of a series of moving averages are strongly correlated, as their formation involved the same members of the original series. It may lead to many moving averages containing cyclic components missing in the original series. This phenomenon is called the Slutsky-Yul effect [59][60][61]. 3. As a method of averaging, in addition to the above-mentioned conventional arithmetic mean can be considered as weighted moving averages, i.e., when the value of the original series in the smoothing interval is summed with specific weights. Such procedures are appropriate if the change in time series over time is nonlinear [59][60][61].
The rationing of time sequences makes it possible to compare the indicators obtained for different objects. In such rationing, the structure of the series (proportionality between levels in the series) remains unchanged. It is possible to compare the calculated indicators and models based on such data [59][60][61].
The criterion of efficiency of smoothing of time series The use of one or another smoothing method depends on meeting the requirements of the problem itself and the series structure. Therefore, it is essential to determine the effectiveness of existing methods. Various criteria are used, among which the most adequate for this problem are: the criterion of turning points and the correlation coefficient between the original and smoothed series [59][60][61].
To assess the smoothing effect, it is proposed to use the criterion of turning points, the content of which is the standard calculation of levels whose values are greater or less than two neighboring [59][60][61]. This criterion is easily implemented in Excel [53].
Correlation coefficient. To estimate the closeness of the relationship between the original -the original series and the smoothed correlation coefficient is used. There is a special option in the add-on "Data Analysis" [53] to find it in Excel.
The formula for calculating turning points
where pi -the value of the price of -periods; wi -the value of the scales for the price of i -periods ago.
The weighted moving average is the arithmetic weighted fluctuations of prices over time [43]. As an analytical tool, it removes some of the shortcomings of conventional sliding but does not eliminate them.
Fig. 13 -Fig. 19 show re-smoothing with Kendall formulas for w = 3, 5, 7, 9, 11, 13, 15 according.
We conducted a comparative analysis of the caloric content of food using various data analysis methods. We used moving average, weighted moving average, linear smoothing, repeated linear smoothing, and exponential smoothing methods to smooth the nutritional value of foods consumed by people with cardiovascular disease [59][60][61]. The relationship between the category of food and its caloric content has been established. Data on the nutritional value of foods consumed by people with cardiovascular disease, collected in the study data [30], allowed to establish the relationship between the category of the product and its caloric content [62][63][64][65]. Studies of the caloric content of food are helpful for patients who suffer from cardiovascular disease or are overweight and want to assess the caloric component of the diet [66][67][68][69][70]. As a result of the comparative analysis, patients and doctors can choose the proper diet to improve patients' health [71][72][73][74][75][76][77][78]. We propose using the approach outlined in [11] to form a diet based on the caloric of food for people with cardiovascular disease. A greater information field creates a stable information field that improves the efficiency of selecting relevant foods and their quantity in the diets of various segments of the population.
To implement recommendations [79] on a healthy living and eating diet, you should choose a diet as soon as possible where you can agree on calorie levels and body weight levels. This recommendation is important when cardiovascular disease occurs or becomes chronic [80][81][82][83][84][85][86][87][88][89][90][91]. We recommend choosing from such groups of products with low (140-220 kJ), low-medium-calorie (220-300 kJ) inform therapeutic nutrition for cardiovascular disease be due to the exclusion of fats and carbohydrates. In addition, it is important to limit the amount of salt and substances that excite the cardiovascular and nervous systems. As can be seen from the results obtained above, the group of products must have medium and low-calorie levels. It includes yesterday's bread; uncomfortable cookies and biscuits; vegetarian soups; lean meats, fish, poultry; milk, sour milk drinks and cheese; dishes from different cereals; boiled pasta; boiled and baked vegetables; soft ripe fruits and berries, honey, jam. It is also important to control the number of micronutrients needed for the heart: potassium and magnesium. Potassium is found in bananas, cabbage, and dried apricots. Magnesium is found in walnuts, carrots, beets and cereals.
With the help of a comparative analysis of the caloric content of food, patients with cardiovascular disease have the opportunity to assess their diet and choose the proper diet correctly. A feature of the proposed approach is harmonizing data on the caloric content of various foods with the body's need to consume these foods.
Problem proper food selection exists for many other diseases, such as congenital or acquired allergies. With the help of the data obtained as a result of our research, people will be able to choose the foods and calories that will promote a healthy diet. Comparative analysis of the caloric content of food will allow people to find an alternative to those products that cause them an allergic reaction, disrupt the proper functioning of the gastrointestinal tract, and more. Comparative analysis of the caloric content of food in the diet is designed to help both doctors and patients with various diseases properly assess their diet and choose a diet that meets the characteristics of the disease, a particular state of the body, and relevant recommendations of doctors on nutrition. Recommended systems based on comparative analysis, cluster analysis and machine learning methods will help to optimize the work of doctors with patients on diet selection.
The problem of proper nutrition is directly related to the quality of food and its caloric content. Many people worldwide suffer from cardiovascular diseases, often caused by being overweight. The paper found that the root cause of the studied problem is the inability of people to properly assess their diet in terms of its caloric content and adequately choose the proper diet.
The comparative analysis of indicators of the caloric content of foodstuff using various methods of the analysis of data is carried out. The methods of moving average, weighted moving average, linear smoothing, repeated linear smoothing, and exponential smoothing of data on the nutritional value of products consumed by people with cardiovascular disease are used. The relationship between the product category and its caloric content has been established.
As a result of research on the caloric content of food, the relationship between the category of product and its caloric content will allow patients suffering from cardiovascular disease or overweight to assess their diet and choose the proper diet to improve their health. An approach to the formation of a diet based on the caloric content of food for people with cardiovascular disease has been proposed, which, with greater availability of data on the caloric content of foods, improves the efficiency of selection of appropriate foods and their number in the diets of different populations. Patients with cardiovascular disease have the opportunity to properly assess their diet and choose the proper diet, the feature of which is to reconcile the data on the caloric content of various foods with the body's need to consume these foods.
Comparative analysis of the caloric content of food in the diet of patients with cardiovascular disease is designed to help both physicians and patients properly assess their diet and choose a diet that meets the characteristics of the disease, body condition, and doctors' recommendations. Problem Proper food selection exists for many other diseases, such as congenital or acquired allergies. With the help of the data obtained as a result of our research, people will be able to choose the foods and calories that will promote a healthy diet. Comparative analysis of the caloric content of food will allow people to find an alternative to those products that cause them an allergic reaction, disrupt the proper functioning of the gastrointestinal tract, and more. We used moving average, weighted moving average, linear smoothing, repeated linear smoothing, and exponential smoothing methods to smooth the nutritional value of foods consumed by people with cardiovascular disease. The relationship between the category of food and its caloric content has been established.