This study examined and created the FDS to help educators and caregivers identify the signs and risk factors of mental health problems in preschoolers. The FDS uses fuzzy logic to identify emotional changes like ADHD, anxiety, and antisocial tendencies in preschoolers’ qualitative behavioural changes. This study used the Preschool Pediatric Symptom Checklist (PPSC)31 to explore preschoolers’ mental health. The PPSC consists of 18 items, including a behavioural and emotional screening instrument created as an important concept for young children’s well-being surveys. These resources are intended to predict preschoolers’ and infants’ behaviour and emotional problems in earlier stages. This study used 18- to 60-month-old children to observe their mental health. The children were screened with 21 items to identify behavioural and emotional problems. The screening process was completed with the help of childcare providers who helped to predict disorders of defiance and problems related to attention deficit hyperactivity disorder (ADHD). The screening process was conducted on three groups of samples: pediatric primary care practice (292 families), referral clinics (354 families), and replication samples (171 families). The parents in the first two groups had 73 drafts of PPSC questionnaires that consisted of items about demographic information, questions about family risk factors, (yes/no) questions about behavioural changes, and questions about anxiety, depression, and problematic behaviour. According to the assessment, PPSC scores are assigned to children to predict their mental health status. During the assessment, if the children are assigned 0, they did not respond at all, 1 means they somewhat responded, and 2 means they very much responded. Suppose the child achieves a total score of 9 or a high score that shows that the child requires further evaluation. The gathered behavioural information about the child is examined using the FDS system to identify mental health decisions. Fuzzy logic offers a conceptual framework that emulates human thinking, making it suitable for accommodating learners’ varied and evolving requirements, behaviours, and interactions. The hybrid approach incorporates machine learning to use data-driven insights, improving decision-making processes’ flexibility and scalability. Due to this synergy, the system can give context-aware recommendations and improve accuracy by analyzing trends in varied datasets. This strategy works well in e-learning because targeted and immediate interventions improve student engagement and performance. Figure 1 outlines the structure of the FDS-based mental health analysis.
Participants were selected from urban and rural regions to ensure that environmental impacts on preschoolers’ mental health were properly represented. Low-income to middle-income families may have had different mental health care accesses. Some parents had just a high school graduation, while others had more academically advanced degrees. Because of this variation, the research included a wide range of parental impacts on children’s mental health. Family members of various racial and ethnic groups comprised the participants, demonstrating the community’s diversity. This variety is needed to understand how cultural influences, including parenting techniques, mental health perspectives, and community resources, affect detecting and treating behavioural and emotional issues in early children. Single-parent homes, parental mental health issues, including anxiety or depression, and economic instability were shown to be important family risk factors. This study know these variables affect kids’ mental health and behaviour. Mental health concerns in parents may cause inconsistent parenting and emotional inaccessibility, which can worsen child behaviour. Single parents may lack time and money to meet their children’s requirements. Financial insecurity may cause stress and limit access to education and healthcare, which can harm young children. Including these risk variables to contextualize children’s behavioural and emotional struggles and appreciate systemic concerns that may affect their mental health. Because of its simplicity, dependability, and emphasis on diagnosing behavioural and emotional issues in young children, the PPSC scoring system was chosen. Using their replies, the method gives kids ratings of 0 to 1 or 2. This permits a more detailed evaluation of symptom intensity and makes the technique straightforward to apply in various contexts. The early diagnosis of at-risk youngsters requires a score of nine or above as a criterion for subsequent examination. This scoring method is especially beneficial when time and resources for comprehensive mental health evaluations may be constrained in pediatric and childcare settings. The research incorporates PPSC scores into the FDS system, using these standardized metrics to provide data-driven insights for mental health decision-making, hence improving the accuracy and effectiveness of the screening process.
This study complies with ethical standards and is approved by the Scientific Research Ethics Committee of the Graduate School in Philippine Women’s University. All methods were performed following relevant guidelines and regulations. This study obtained the informed consent of all human participants. For children, permission was obtained from the guardian.
Structure of the FDS-based mental health analysis of preschoolers.
The FDS system
Generally, medical science and knowledge have several uncertainties when analyzing people’s health. In particular, when examining preschoolers’ mental health, there are various uncertainties, such as limited expressive abilities, developmental variability, external influences, and limited self-awareness. Limited expressive abilities lead to struggling while expressing one’s emotions and feelings, which creates anxiety and depression. In addition, young children are easily influenced by cultural factors, family dynamics, and socioeconomic conditions. These factors affect children’s mental health and are difficult to separate from those impacting typically developing children. As such, FDSs have been applied in medical science to address uncertainty issues. The fuzzy approach considers every feature in medical applications and provides recommendations effectively. The FDS is defined according to the membership function with a 0 to 1 interval. The defined membership value quantifies the input and visually represents the fuzzy set. For fuzzily, every member is represented by a membership rank used to classify the members into a particular class. The membership function is defined with a closed interval \(\:\left[\text{0,1}\right],\) and the members are defined as the degree of membership in the set.
Assuming that the reference set is \(\:\mathcal{H}\), the fuzzy set \(\:\mathcal{F}\) is defined on \(\:\mathcal{H},\) and the membership function is represented as \(\:{\mu\:}_{A}\left(x\right)\). \(\:{\mu\:}_{A}\left(x\right)\) is defined from \(\:\mathcal{H}\) to \(\:\left[\text{0,1}\right]\). Then, the membership of \(\:A\) elements of \(\:\mathcal{H}\) in set \(\:\mathcal{F}\) is defined using Eq. (1).
$$\:\mathcal{F}=\left\{\left(x,{\mu\:}_{A}\left(x\right)\right):{\mu\:}_{A}\left(x\right):\mathcal{H}\to\:\left[\text{0,1}\right]\right\}$$
(1)
Trapezoidal and triangular membership functions are utilized during the analysis process of mental health decisions, as defined in Eq. (2).
$$\:{\mu\:}_{A}\left(x\right)=\left\{\begin{array}{c}\frac{x-\alpha\:}{m-\alpha\:}\:\:\:\:\alpha\:
(2)
Equation (2) is the triangular membership function applied to the fuzzy set when the membership function appears at only one point. Here, x is the input, and (\(\:\alpha\:,\:m,\:b)\:\) is the parameter in the triangular function. Then, the trapezoidal function is computed using Eq. (3), employed in the set when the maximum membership appears in one interval.
$$\:{\mu\:}_{A}\left(x\right)=\left\{\begin{array}{c}\frac{x-\alpha\:}{m-\alpha\:}\:\:\:\:\alpha\:
(3)
In addition, the triangular and trapezoidal membership and non-membership functions must be computed while exploring every input of preschoolers. The non-membership function is estimated using Eq. (4).
The computed \(\:{\nu\:}_{A}\left(x\right)\) ensures the degree of non-membership value of \(\:\mathcal{H}\) to \(\:A\), and it must satisfy two conditions: \(\:\left\{\left({\mu\:}_{A}\left(x\right),{\nu\:}_{A}\left(x\right)\right)\in\:\left[\text{0,1}\right]\right\}\) and \(\:\left\{\left({\mu\:}_{A}\left(x\right)+{\nu\:}_{A}\left(x\right)\right)\le\:1\right\}\). In the FDS, the fuzzy numbers of the degrees \(\:{\mu\:}_{A}\left(x\right)\) are 0 and 1; the degree of the \(\:{\nu\:}_{A}\left(x\right)\) value is the complement of the \(\:{\mu\:}_{A}\left(x\right)\) value. The combination of \(\:{\mu\:}_{A}\left(x\right)\) and \(\:{\nu\:}_{A}\left(x\right)\) creates complexity while handling decision-making. Thus, fuzzy set components such as fuzzy rule databases, fuzzifiers, inference engines, and de-fuzzifiers are utilized to improve decision-making. The overall components of the FDS are illustrated in Fig. 2.
Fuzzification and De-fuzzification Process.
Figure 2 illustrates the components of the FDS system, which include fuzzification, fuzzy rule base generation, an inference engine, and de-fuzzification. These components reduce decision-making complexity while analyzing preschoolers’ mental health. The fuzzification receives input X from the data sources that have to be converted into fuzzy set \(\:\mathcal{F}\) and the degree of membership \(\:{\mu\:}_{A}\left(x\right)\). Then, fuzzy rules are generated according to the expert’s knowledge and used to map the input to the membership value. The mapping process P is performed in the inference engine. Finally, de-fuzzification is performed to convert the fuzzy output into a crisp output (CO), which helps to provide recommendations, classifications, and decisions for specific applications.
First, fuzzy sets are created using clean, real-world input data (such as behavioural ratings from the Preschool Pediatric Symptom Checklist) inside the framework of the Fuzzy Decision Support (FDS) system. This is accomplished by using membership functions that specify the extent to which each input falls into a certain fuzzy category (e.g., low, medium, or high) according to established thresholds. For example, according to the membership function, a behavioural score of 8 may be categorized as moderate,’ with a membership degree of 0.7 falling into that category and 0.3 into the ‘high,’ respectively. The system can interpret the fuzzy inputs better because they reflect the inherent ambiguity and imprecision in the real-world data. The defuzzification technique is then used to make decisions based on a clear value rather than a fuzzy one. Common techniques for this include the centroid and the mean of maxima approach. A fuzzy set depicting the severity level or the required intervention is produced by combining the fuzzy results from the inference system using fuzzy rules. This process may be used for many behavioural disorders, such as anxiety or aggressiveness. The defuzzification process produces a single, usable result by calculating the weighted average of the values in the fuzzy set. For instance, to ascertain the necessary degree of intervention, a fuzzy output suggesting the need for a gentle intervention may be transformed into a specific suggestion, such as a score of 7 out of 10.
Fuzzification
Fuzzification is an important step in the FDS system in which the crisp input is converted into a fuzzy set, and the degree of membership is determined. The membership value is [0, 1]; if 0, the input does not fall under the fuzzy set; otherwise, it belongs to the fuzzy set. Here, the mapping process is defined as \(\:{x}^{*}\in\:U\subset\:{R}^{n}\) to the fuzzy set \(\:\mathcal{F}\)in \(\:U\). During this process, the triangular fuzzy generator reduces complex computations and perturbations.
$$\:{\mu\:}_{\mathcal{F}}\left(x\right)=\left\{\left(1-\frac{\left|{x}_{1}-{x}_{1}^{\text{*}}\right|}{{b}_{1}}\right)\widehat{\mathcal{H}}\dots\:\dots\:\widehat{\mathcal{H}}\left(1-\frac{\left|{x}_{n}-{x}_{n}^{\text{*}}\right|}{{b}_{n}}\right)\right.\left|{x}_{1}-{x}_{1}^{\text{*}}\right|\le\:{b}_{i}$$
(5)
In Eq. (5), the positive parameter is denoted as \(\:{b}_{i}\), and t-soft is represented as \(\:\widehat{\mathcal{H}}\), which is chosen here as a minimum. According to the fuzzification process, the input is converted with the help of the membership function saved in the fuzzy knowledge base. The fuzzification process is applied to preschoolers’ mental health analysis. The fuzzification process defines the fuzzy set and membership function. According to the analysis, the input is a social engagement fuzzy set (SEF)\(\:\{SEF:low,\:moderate\:and\:high\}\). For the low category fuzzy set, the triangular membership function takes the parameters (0, 0, 30); the moderate fuzzy set has the parameters (10, 50, 70); and the high fuzzy set takes the parameters (80, 100, 100). The membership value of the SEF is defined in Fig. 3.
Figures 3(a) and (b) represent the fuzzification process based on a fuzzy set of low and moderate social engagement levels. Every fuzzy set is linked with the membership function (triangular), which denotes the children’s response-related degree of the membership function concerning social engagement. Figure 3(a) shows that the membership value is minimal in the low fuzzy set when the SEF value is 0 and progressively increases until it reaches 20. For each entry with a low SEF, the \(\:{\mu\:}_{A}\left(x\right)\) value is estimated by checking the conditions \(\:\alpha\:
Membership values. (a) low-SEF, (b) moderate-SEF, (c) high SEF.
Figure 3(c) indicates that a high fuzzy set value is related to membership function analysis, in which the degree of membership is minimized to 80 and increases gradually when it reaches 100. The computed \(\:{\mu\:}_{A}\left(x\right)\) value is denoted as the preschoolers’ fuzzification process of social engagement. According to the analysis, social engagement is not always accurate to low, high, and moderate, which can be between because of various uncertainties and factors. Table 1 portrays the computations of low, moderate, and high-value-related \(\:{\mu\:}_{A}\left(x\right)\)-generalization estimates.
The degree of membership quantifies the level to which each input value is an element of the given fuzzy set. The mathematical Equation is derived from triangle membership functions and their associated parameters. Fuzzification enables the expression of vague data and enhances the decision-making process when uncertainties are present. The fuzzy sets obtained and their corresponding degrees of membership can be subsequently employed in fuzzy inference systems to facilitate decision-making. Hence, the fuzzification process assesses how children’s health information changes into a fuzzy processing format input. Likewise, children’s mental health is assessed by various factors, such as activity level, eating pattern, sleep duration, the ability to focus, and sociability. The fuzzy system uses intelligible language rules that exploit imprecise, qualitative characteristics to yield indications regarding a child’s cognitive and emotional development that necessitate support. The process of graduate fuzzification enables the integration of imprecise, real-world notions about human health and behaviour. Figure 4 outlines the fuzzification process of various factors and their respective outcomes.
Analysis of the fuzzified output of mental health-related factors.
According to Fig. 4, the crisp inputs are taken from different activity levels and fed into the triangular membership function, which is utilized to predict fuzzified output values such as low, moderate, and high. The inputs are mapped with the membership function to obtain the degree of membership. The obtained inputs are passed on to fuzzy rules to make effective decisions about a child’s mental health.
Fuzzy rule generation
The next important step is fuzzy rule generation, framed using the “if-then” condition. The fuzzy set rule is the crucial FDS step framed according to the expert’s knowledge. According to the PPSC’s resources, the fuzzy rules are framed to identify the child’s mental health. The generated rules are useful in creating a positive environment for children to improve their lives. Here, a few fuzzy rules are listed in Table 2.
According to fuzzy rules, the output fuzzy set is derived as a low-, moderate-, or high-risk list. The generated fuzzy set is utilized in clinical analysis to categorize preschoolers’ mental health effectively. Then, the fuzzy inference engine employs the fuzzy set to identify the output, which helps to determine the child’s mental health. The fuzzy set activity levels (low, moderate, and high), eating patterns (irregular, somewhat regular, and regular), ability to focus (distracted, inconsistent, and attentive), sleep duration (low, sufficient, and high), and sociability (introverted, ambivert, and extroverted) are the inputs to the fuzzy inference engine. The rules are framed by a combination of inputs that help to determine the child’s mental behaviour output. Here, the output of the ability to focus is obtained by considering sleep duration and activity level. Likewise, anxiety and perfectionism are combined to frame the rules to obtain the output value. Here, the fuzzy set is \(\:Activity\:Level\::\{low,\:moderate,\:high\}\), \(\:Eating\:pattern:\:\{irregular,\:somewhat\:regular,\:regular\}\), and \(\:Ability\:to\:focus:\:\:\{distracted,\:inconsistent,\:attentive\}\). For the fuzzy set, the fuzzy rules are framed with the help of expert knowledge. The rule is “If the activity level is moderate and the eating pattern is somewhat regular, then the ability to focus is inconsistent.”
According to the rules and criteria, the triangular membership function is applied to the parameters to obtain the output for the input. Let us assume that the activity level has a fuzzy set: low (0, 0, 30), moderate (20, 50, 80), and high (70, 100, 100). The eating pattern fuzzy set is irregular (0, 0, 30), somewhat regular (20, 50, 80), or regular (70, 100, 100). The ability to focus is set as distracted (0, 0, 30), inconsistent (20, 50, 80), or attentive (70, 100, 100). For the entire fuzzy set, the degree of membership is estimated for the above-defined fuzzy rule. Here, the activity level is 60, and the eating pattern is 60. The activity level of the membership function is moderate; it has a value of 0.33, according to Eq. (2). The degree of eating pattern of the membership function is somewhat regular at 0.33. Then, the fuzzy inference system is computed using the minimum value, which is defined in Eq. (6):
$$\:Inference=\left.\begin{array}{c}\text{min}\left(\mu\:\left(moderate\right),\:\mu\:\left(somewhat\:regular\right)\right)\\\:\text{min}\left(\text{0.33,0.33}\right)\\\:0.33\end{array}\right\}$$
(6)
Then, the ability to focus is the degree to which the membership function is inconsistent, 0.5. During the computation, the degrees of activity level of the membership and eating pattern values are fed into the AND operation to identify ability-related mental problems. The ability to focus on the final output is estimated by de-fuzzifying and aggregating the rule outputs; Fig. 5 shows the graphical structure of the fuzzy inference process concerning the rule.
Figure 5 depicts the deduction from the given fuzzy rule and membership functions. The rule’s degree of membership is highest in the region when the activity level and eating pattern have moderate values. According to the fuzzy output, a positive environment is created for preschoolers by considering various factors such as the ability to focus, the eating pattern, and the activity level.
De-fuzzification process.
A positive environment for preschoolers
The fuzzy rules are created with the help of expert knowledge, which covers all possible solutions for preschoolers to maintain their mental health. The child’s moderate activity level is encouraged by providing physical activities, outdoor activities, movement, encouraged games, and structured playtimes according to the child’s age. Then, eating patterns are regularized by creating a structured, consistent mealtime routine. In addition, a social and positive atmosphere and balanced, healthy foods are also provided. Focus is maintained by designing play and learning areas with minimum distraction. Then, activities are broken into acceptable segments that help the preschooler understand and pay attention. The unique variations in activity levels, food habits, and the ability to focus among individuals should be acknowledged and honoured. Activities and routines should be customized to meet every child’s distinct requirements and inclinations. Commendation and constructive reinforcement are provided for desired actions, such as sharing, collaboration, and attentiveness. A reward system should be implemented that prioritizes good acts to incentivize and strengthen positive behaviour. The physical environment should be secure, engaging, and favourable for exploration. A feeling of safety and confidence can be developed by upholding regular schedules and explicit standards. A transparent way to communicate with parents should be developed to obtain valuable information about a child’s conduct at home. Teachers should cooperate with parents to establish uniformity across domestic and preschool settings. If the child is distracted and cannot focus, the learning environment and schedule should be structured to support attention. A proper routine and break should occur between every activity to improve attention. Then, techniques to calm down, such as quiet reading areas, fidget toys, and breathing processes, should be used to maximize preschoolers’ health status. The effective observation and assessment of children’s behaviour helps to predict their mental health according to the fuzzy set, membership function, and rules. Thus, the objective of reducing the causes of discomfort while promoting the growth of emotional, behavioural, and social abilities has been successfully established. Implementing a caring, supportive, and customized approach will give preschoolers a sense of security, involvement, and positivity.
link