Prototype Data Mining Pola Jabatan Fungsional Dosen Menggunakan Teknik Emerging Pattern: Studi Kasus Universitas Mercu Buana

Abstract

Abstract
 
 Lecture Functional position is used to ensure the development of lecturer carrier, structural position, and professionalism improvement. Every lecturer has a right to propose a functional promotion as long as their education background is linear. Data mining has been widely used to analyze data into information in term of patterns that are easily understand by the users. In order to create a functional position pattern that produces beneficial and useful results, the data mining techniques are utilized for analyzing, for example the emerging pattern. The emerging pattern technique represents very strong distinguishing knowledge between datasets and shows an accurate classification ability. The results of data mining with emerging pattern techniques were used in this study to obtain patterns of functional positions and to classify lecturers' educational linearity. The learning data results obtained a confidence value for the sex dataset confidence values for male and female itemset of 50%. Dataset functional lecturers having higher confidence was Lectors of 300 points with a confidence value of 73%. Department dataset which had higher confidence value was psychology by 85%. The highest confidence value in the dataset from tertiary institutions originates from abroad by 62% and for the age dataset that most have functional lecturer positions were aged 51-70 years with a confidence value of 55%.
 
 Keywords: Data Mining, Emerging Pattern, Fungsional Position, Lecturer, Linear Education
 
 Abstrak
 
 Suatu pola untuk menjamin pembinaan karier kepangkatan, jabatan dan peningkatan profesionalisme dosen disebut Jabatan fungsional dosen. Setiap dosen berhak untuk mengajukan kenaikan jabatan fungsional dengan syarat salah satunya adalah liniear dalam bidang ilmu. Data mining mampu menganalisis data menjadi informasi berupa pola yang mempunyai arti bagi pendukung keputusan. Agar proses dalam pencarian pola jabatan fungsional ini menghasilkan nilai tambah dan berguna maka dibutuhkan teknik data mining untuk menganalisanya. Salah satu teknik dalam data mining adalah emerging pattern. Teknik emerging pattern merepresentasikan pengetahuan pembeda yang sangat kuat antara dataset dan menunjukkan kemampuan klasifikasi yang akurat. Hasil dari data mining dengan teknik emerging pattern adalah mendapatkan pola (pattern) dalam jabatan fungsional dan mengelompokkan kelinieritasan pendidikan dosen. Dari hasil learning data diperoleh nilai confidence untuk dataset jenis kelamin nilai confidence untuk itemset Laki-laki dan perempuan sebesar 50%. Dataset jabatan fungsional dosen yang nilai confidence nya lebih besar adalah Lektor 300 dengan nilai confidence sebesar 73%. Dataset jurusan yang mempunyai nilai confidence terbesar adalah psikologi sebesar 85%. Nilai confidencen yang paling besar pada dataset asal perguruan tinggi berasal dari luar negeri sebesar 62% dan pada dataset usia yang paling banyak memiliki jabatan fungisonal dosen berada pada usia 51-70 tahun dengan nilai confidence sebesar 55%.
 
 Kata Kunci: Data Mining, Emerging Pattern, Jabatan Fungsional, Dosen, Linier.

Application of Classification Data Mining Technique for Pattern Analysis of Student Graduation Data with Emerging Pattern Method

“…The mining results contain the support value, growth rate (GR), and confidence. Based on research [12], the support value is used to see the tendency of data, whether it is more dominant to dataset 1 or dataset 2. The growth rate (GR) value is used to see class differences.…”

“…This Emerging Pattern (EP) captures significant changes and differences between datasets. Emerging Pattern (EP), which has a large growth rate (GR) value, can distinguish the characteristics of two datasets and can build very interesting classifiers [12]. According to research [13], the Emerging Pattern method is divided into discriminating between two datasets and classifying more than two datasets.…”

“…JEPs (jumping emerging patterns) occur when the support value of dataset 1 or as the denominator in equation ( 2) is 0 so that the growth rate (GR) will be (∞). Because the growth rate (GR) is worth (∞), itemsets containing Jumping Emerging Patterns have no confidence value [12].…”