Accurately Determining Labor Test Results Using the Rough Set Method

: An exam is something that must be done to test a person's ability or intelligence. The laboratory exam in the Computer Systems study program at Putra Indonesia University "YPTK" Padang consists of a digital systems exam, a fuzzy logic control exam


Introduction
Machine learning is one method used to solve problems in the environment.In machine learning (Alkinani et al., 2020;Gao & Wu, 2020), there is a more detailed part in decision-making, namely data mining.In data mining, there is a rough set method that can be used to assist decision-making (Chinnaswamy & Srinivasan, 2017;Kurniawan et al., 2018).An exam is something that is used to test a person's abilities or learning results, A laboratory test must be carried out by students before taking the Comprehensive exam (Pelton, 2017).The laboratory exam in the Computer Systems study program consists of 3 parts, namely the Digital Systems Exam, Fuzzy Logic Control, and Tool Presentation.So far, the process of determining student laboratory exam results has been carried out manually (González-Calatayud et al., 2021;Swiecki et al., 2022;Abdulrahman et al., 2020).This process results in decision-making taking a long time.In this research, the rough set method was used to help make laboratory exam decisions for students who had registered to take the laboratory exam (Attaullah et al., 2023;Chen et al., 2023;Hariri et al., 2019).Wang et al. (2023) and Puška et al. (2023) stated Rough Set is a method for dealing with ambiguity and uncertainty introduced in the processing of imprecise information.

Figure 1. Solution algorithm using the rough method set
The settlement scheme using the Rough Set method consists of several stages, namely the Decision System, Equivalence Class, Dicernibility Matrix, Dicernibility Matrix Modulo D, Reduction, and Knowledge (Zuhdi, 2022).This research aims to accurately determine the results of the digital system laboratory exam, fuzzy logic, and the percentage of tools that students pass or fail.

Method
The method used in this research is the Rough Set method (Raharjo & Windarto, 2021;Halder et al., 2019) and to simplify the methodology and system design process, an analysis and design flow chart can be created as shown in the image below:

Result and Discussion
Rough set theory is a tool mathematics to deal with disadvantages clarity and uncertainty introduced to process the absence uncertainty and inaccurate information (Qu et al., 2020;Pendrill, 2014;Demin, 2020).Rough the set has been widely applied in many ways real problems in medicine, pharmacology, engineering, banking, finance, market analysis, environmental management and etc (Kocornik-Mina et al., 2021;Khairunnessa et al., 2021).According to Dagdia et al. (2020), Swiniarski et al. (2003), and Manurung et al. (2018), that, stages in use The Rough Set algorithm is as follows: Data selection (Selection of data will be used); Establishment of a Decision System contains condition and attribute attributes decision; Establishment of Equivalence Class, namely by eliminating data which is repetitive; Formation of the Discernibility Matrix; Modulo D, namely the matrix contains comparisons between data different condition attributes and attributes decision; Produce reduct with using boolean algebra; Produce rules (knowledge) (Sianturi et al., 2021;Nurhidayat et al., 2020).
Rough Set was created by Zdzislaw Pawlak in the early 1980s, in order to mathematically reveal the concept of vagueness, its main goal is to be an automated process of transforming data into knowledge (Pięta et al., 2019;Pięta & Szmuc, 2021).Rough sets are a mathematical approach to knowledge that is not perfect, this is important in fuzzy logic (Slim & Nadeau, 2020;Bobillo & Straccia, 2012).Rough set lies in the fact that, based on a set of objects, a set attributes and decision values, one can create a rule to find upper and lower estimates, and the boundary region of the set object (Herbert & Yao, 2009;Del Giudice et al., 2017).After the rule is created, new objects can be created easily classified into one of the regions (region) (Sarker, 2021).The concept of rough sets in general can be defined by 2 topologies, namely interior and closure (Ali et al., 2013;AL-Khafaji & Hussan, 2018).
The basic idea of Rough Set (RS) is a mathematical technique used to handle problems of uncertainty, imprecision and ambiguity in Artificial applications Intelligence (AI) (Liu et al., 2022;Kristanto et al., 2021).RS is related to the classification from the table.Even in theory RS is related to discrete data, RS is usually used in conjunction with engineering another to perform discreetization on the dataset (Ali et al., 2023;Zhang et al., 2020;Ayub et al., 2022).The main features of RS data analysis are non-invasive, and the ability to handle qualitative data.The results of the RS analysis can be obtained used in the Data Mining and Knowledge Discover processes.In this research, several stages were carried out to obtain the desired results.

Decision Systems
Data is prepared in table form containing Condition Attributes and Decision Attributes.Condition attributes are placed in the left column, while decision attributes are in the right column.Condition attributes consist of 1 or more attributes while decision attributes only consist of 1 attribute.The decision system table can be seen in the table below.

Conclusion
The Rough Set method can help provide accurate decisions on student laboratory exam results.This research produces 29 rules as knowledge, namely {Digital Systems} or {A} = 3 rules, {Fuzzy Logic} or {B} = 3 rules, {Tool Presentation} or {C} = 3 rules, {Fuzzy Logic, Tool Percentage} Or {BC} = 6 rules, {Digital System, Fuzzy Logic} Or {AB} = 6 rules and {Digital System, Tool Percentage} Or {AC} = 8 rules.

Figure 2 .
Figure 2. Analysis flow chart In this research, 20 students took laboratory exams, namely the Digital Systems, Fuzzy Logic Control, and Tool Percentage exams.

Figure 3 .
Figure 3. Distinction Matrix Dicernibility Matrix Modulo D The columns in the Matrix are filled with a set of Condition Attributes that have different Condition values and also different decision values, which can be seen in the figure 4.

Figure 7 .
Figure 7. Labor exam data in microsoft excel Dynamic Reduct from research conducted can seen in the figure 8.

Figure 8 .
Figure 8. Dynamic reductThe resulting reduction from the data in the research is shown in Figure9.

Table 1 .
Decision System Equivalence Class is a grouping of objects that have the same Condition Attribute values, which can be seen in the table below.

Table 2 .
Equivalence Class is a Grouping of Objects that Have the Same Condition Attribute Values Discernibility MatrixThe columns in the Matrix are filled with a set of Condition Attributes that have different Condition values, which can be seen in the figure below.

Table 3 .
The Rules that are Formed from the Data Provided in a Study this as Many as 29 Rules