Page 188

www.rsisinternational.org

MIC3ST 2025 | International Journal of Research and Innovation in Social Science (IJRISS)

Virtual Conference on Melaka International Social Sciences, Science and Technology 2025

ISSN: 2454-6186 | DOI: 10.47772/IJRISS | Special Issue | Volume IX Issue XXIII October 2025

difficulty to record the data. Therefore, diabetes patients need to manage their insulin administration timely. It

depends on several factors such as carbohydrate intake, current blood glucose levels and individual

preferences. To address this issue, this project will focus on developing a expert system using rule-based

forward chaining technique that make use of Insulin-for-carb table to help diabetes patients in calculating the

appropriate insulin dose for each meal. The system will also determine the carbohydrate content and

recommend the appropriate insulin dosage if users record the foods and beverages they consume. The system

also monitors insulin dosages and blood sugar levels. These users’ feedback showed the similarity of

healthcare professionals’ practice. As conclusions, this system is reliable to enhance diabetes management and

help in reducing the risk of both hyperglycemia and hypoglycaemia.

before and after eating four to seven times a day. Diabetic patients are advised to consume nutritious food with

proper portion size. They also need to know the carbohydrates in the food taken to calculate the amount of

insulin needed.

Based on interview with one of doctors in Dr Akmal Saleh Clinic, Dr Muhamad Hafiz Bin Abdullah, diabetes

mellitus (DM) was a condition when the blood contained excessive amounts of glucose (sugar) as a result of

inadequate or inefficient insulin. It resulted from insufficient production of the hormone insulin, which was

generated by the pancreas, an organ in the body. The function of insulin was to facilitate the transfer of glucose

from the bloodstream into the body's cells, hence regulating the body's glucose balance. This statement was be

supported by Dottie (2014) that stated when insulin was either insufficient or ineffective, the blood glucose

(sugar) level rises, resulting a disease known as diabetes mellitus. The normal range for random blood glucose

Page 189

www.rsisinternational.org

is 7.8 mmol/L – 11.1 mmol/L. Furthermore, Dr. Hafiz explained that the treatments for diabetes patients were

diet control, insulin and oral hypoglycemia medication. Mostly, insulin injections were given to diabetes

patients Type 1. There are five types of insulin such as rapid-acting insulin, short-acting insulin, intermediate-

acting insulin, mixed insulin and long-acting insulin. This statement was supported by insights from Nurul

Hazwani Izni Binti Mohd Nizam, one of diabetic patient interviewed to gain an understanding of the process

followed when administering insulin.

Figure 1: Glucometer monitoring chart

Diabetic patients who need to consume insulin, have to check their blood sugar themselves, and record it

calculating the insulin dose, and have inconsistencies in tracking their blood glucose level as survey done in

Figure 2. Therefore, this paper addresses these issues by developing a system will calculating and determine

the appropriate insulin dose using the Rule-Based Forward Chaining technique.

Diabetes is a collection of metabolic illnesses characterized by hyperglycemia caused by insulin secretion or

action abnormalities, which result in long-term organ damage and dysfunction (Basevi, V., 2011). Based on

Arneth et al., (2019), Type 1 and Type 2 diabetes are chronic diseases that afflict nearly 425 million people

globally, causing changes in metabolites such as glucose, fructose, amino acids, and lipids

Page 190

www.rsisinternational.org

According to Brutsaert (2023), a common disease called diabetes mellitus is caused by the body's inability to

control blood glucose levels, which causes the body to produce an excessive amount of urine. This occurs as a

result of the body producing insufficient insulin or using it improperly. The World Health Organization

(WHO) reports that the number of people with diabetes increased from 108 million in 1980 to 422 million in

2014, with low- and middle-income countries experiencing a disproportionate increase in cases compared to

high income countries. The incidence of the condition is still rising, and it is becoming a major cause of death

worldwide. In 2019, diabetes was directly responsible for almost 1.5 million deaths (Antar et al., 2023).

(DiCARE) (2006-2008) study, 71.8% of children under 20 years old with diabetes mellitus had Type 1

diabetes. The median age of diagnosis was 7.6 years (IQR 4.6-10.8), with the majority (58.3%) presenting with

diabetic ketoacidosis. Early detection of T1DM in children is crucial, especially before the onset of Diabetic

ketoacidosis (DKA), which can lead to severe morbidity and mortality (Hong et al., 2017). Meanwhile, Type 2

Diabetes Mellitus (T2DM), usually affects adults and is brought on by insufficient or resistant insulin

production in the body. According to Endocrine (2021), T2DM accounts for more than 90% of adult-onset

diabetes cases in Malaysia. Diabetes and prediabetes can coexist with other noncommunicable diseases.

T2DM, characterized by hypertension, dyslipidemia, and obesity, has a high worldwide and local prevalence

and poses a significant socio-economic burden due to vascular consequences and premature death.

diabetes in order to address glucose excursions during mealtimes. The insulin's pharmacology in relation to the

meal's glucose profile. They must use insulin injections administered by syringes, insulin pens, or insulin

pumps to control their condition. Modern insulin pumps continuously provide basal insulin by using an insulin

analogue with a fast action. Bolus calculators for insulin pumps also determine the required units of insulin

depending on food consumption, active insulin time, and parameters.

According to Al-Mafrji et al. (2023) expert systems is the Knowledge-Based Systems (KBIS), a subset of

Computer-Based Systems (CBIS) that studies the design and development of human-like computer systems

with intelligence. Knowledge-based information systems can be used as an expert consultant for end users tally

with its complexity. Therefore, knowledge-based information systems are widely used in various fields such as

and their attributes. Secondly, user interfaces allow communication between the user and the expert system. It

enables users to input queries and receive response (Al-Mafrji et al., 2023). Thirdly, the inference engine

applies inference rules to the knowledge base to make a conclusion as it is the main processing unit of the

Page 191

www.rsisinternational.org

system. There are two inference engine which are forward chaining and backward chaining (Al-Mafrji et al.,

2023). According to Anwar (2023), the method begins with the established rules and facts, then adds the

conclusion to the known facts by using the inference rules is forward chaining. Meanwhile, backward chaining

is the technique of backward reasoning develops the known facts first and works backward to reach the goal.

Lastly, according to Janjanam et al., (2021) the working memory, output memory and interpreter is the

components in Blackboard Architecture framework

Rule Based Systems (RBS) are computer programs from research into artificial intelligence. This system

performs activities that are often performed by human experts by imitating human knowledge and behaviour

(Anwar, 2023). The rule base and the inference engine are the two main parts of a rule-based system. The

inference engine uses the facts in the rule-based to help generate new knowledge based on the facts. There are

two types of rule-based technique such as forward and backward chaining techniques can be used, and the IF-

THEN rule is frequently used in this process (Masri et al., 2019).

Based on Rizki Gunawan Tanjung (2023), forward chaining begins with existing data and progresses to a

conclusion by applying rules and deriving new information step by step. This method is particularly beneficial

for solving problems that begin with preliminary information and continue to a final answer in a sequential

manner. The inference engine evaluates rules and executes actions based on user input that meets the

conditions (Jalinus, 2019).

According to Ramadhani et al., (2021), using a set of rules for behaviour is known as forward chaining. This

approach involves selecting which rules to execute based on the data, and then executing the rules. Until a

result is achieved, the procedure adds information to working memory. Typically, decision-makers receive

support from the expert knowledge base based on the findings of their fact-finding efforts. The knowledge

base representation used is known as rule-based and is expressed as statements to convey the issue of essential

Page 192

www.rsisinternational.org

THEN correction insulin dose is needed

Page 193

www.rsisinternational.org

specific ICR. The Knowledge Base Table: Insulin Carb Ratio stored the ICR values for different patients,

allowing the system to calculate insulin doses according to each individual’s ratio

Page 194

www.rsisinternational.org

However, the system entered conflict resolution mode when multiple rules activated simultaneously. The

system used conflict resolution strategies to determine which rules should take precedence when multiple rules

were active. The system applied a rule selection process that used both rule entry order and rule specificity to

determine which rule would be executed first. Lastly, the system generated its output through the execution of

rules during the final stage. The system generated the output which was insulin doses for carbohydrate intake.

a) Carbohydrate Intake (grams): Specified amount of carbohydrate consumed.