scholarly journals PREDIKSI TINGKAT KEBERHASILAN STUDI KINERJA SANTRI MENGGUNAKAN ALGORITMA C 5.0

SAINTEKBU ◽  
2020 ◽  
Vol 12 (2) ◽  
pp. 45-57
Author(s):  
Achmad Agus Athok Miftachuddin
Keyword(s):  
Data Mining ◽  
Tracer Study

Keberhasilan lembaga pendidikan pesantren dapat diukur dari keberhasilan santrinya. Dengan memprediksi kemungkinan hasil dari proses pembelajaran berdasarkan hasil prediksi dapat membantu suatu lembaga pendidikan pesantren, dengan menyesuaikan faktor-faktor yang berkontribusi dan mempengaruhi tingkat keberhasilan studi kinerja santri. Dan dengan memanfaatkan teknik data mining yang dapat digunakan untuk meningkatkan tingkat keberhasilan dan mengurangi kegagalan santri. hal ini dapat sangat membantu lembaga pendidikan pesantren untuk meningkatkan kecakapan lulusannya, karena data mining merupakan solusi terbaik untuk menemukan pola tersembunyi dan dapat memprediksi tingkat keberhasilan studi kinerja santri. Penelitian ini menyajikan model berdasarkan pohon keputusan klasifikai algoritma C 4.5 yang digunakan dalam model ini dengan tracer study online yang diisi oleh alumni santri, tracer study online ini terdiri dari 12 Pertanyaan yang mencakup bidang, seperti sekolah, status sekolah, jumlah saudara, riwayat sebelum dipondok, kategori pondok, jarak, dan beasiswa merupakan data status sosial santri ketika masih menjadi calon santri yang akan menentukan sekolah menengah atas. Dan data riwayat akademik terdahulu santri yang meliputi rata-rata raport, rata-rata usbn, rata-rata un, prestasi dan level keberhasilan yang paling terkait dan mempengaruhi studi kinerja santri. Dan sebanyak 300 tracer study alumni di kumpulkan dan diimplementasikan kedalam bahasa pemrograman r untuk membangun model ini menggunakan algoritma C 4.5. Dari hasil pengujian menggunakan model tersebut diperoleh akurasi sebesar ............. dan dapat disimpulkan bahwa algoritma C 4.5

Rekomendasi Grup Pada Website Alumni Teknik Informatika Universitas Muhammadiyah Malang

Repositor ◽  
2020 ◽  
Vol 2 (12) ◽  
pp. 1647
Author(s):  
Hermansyah Adi Saputra ◽  
Galih Wasis Wicaksono ◽  
Yufis Azhar
Keyword(s):  
Data Mining ◽  
Information System ◽  
Information Systems ◽  
Tracer Study ◽  
Group Recommendation ◽  
Number Of Clusters ◽  
Average Value ◽  
Exchange Information ◽  
Group Object ◽  
Almost All

AbstrakBelakangan ini hampir seluruh universitas yang ada di indonesia memiliki sistem informasi alumninya sendiri-sendiri. Sistem informasi alumni mampu memberikan informasi tentang kondisi alumninya setelah menyelesaikan masa perkuliahannya. Alumni merupakan aktor yang berperan penting dalam pendidikan. Saat ini jurusan Informatika Fakultas Teknik Universitas Muhammadiyah Malang telah memiliki website alumni. Permasalahannya belum adanya sistem yang memberikan alumni rekomendasi grup pada sistem, sehingga para alumni mampu saling bertukar informasi didalamnya. Dengan adanya data alumni dan juga di dukung dengan adanya tracer study, dapat di bentuk suatu rekomendasi grup dari data tracer study. K-medoid adalah metode pengelompokan data ke  dalam  sejumlah cluster  tanpa  adanya  struktur  hirarki antara satu dengan yang lainnya. Algoritma k-medoid memiliki nilai coefficient yang lebih tinggi di bandingkan dengan k-means dalam penelitian ini. Yang mana k-medoid mendapatkan nilai rata-rata Silhouette Score 0.7325888099 dalam pengujian dengan jumlah cluster 5 dan perulangan sebanyak 10 kali. Jika dibandingkan dengan k-means yang hanya memiliki nilai rata-rata Silhouette Score 0.6872873866.AbstractLately, Almost all universities in Indonesia have their own alumni information systems. The alumni information system is able to provide information about the condition of its alumni after collage graduation. Alumni are actors who play important role in education. Currently, the Department of Informatics, Faculty of Engineering, University of Muhammadiyah Malang has an alumni website.  The problem is the absence of system that gives alumni group recommendation on the system, so that alumni are able to exchange information in this website. With the alumni data and also supported by the existence of a tracer study, it can be formed as group recommendation from the data tracer study. Clustering is one of tools in data mining that aims to group object into clusters. K-medoid is a method of grouping data into a number of clusters without hierarchical structure from one another. The k-medoid algorithm has higher coefficient value compared to k-means in this study. This K-medoid gets an average value of Silhouette Score 0.7325888099 in testing with the number of clusters 5 and repetitions 10 times. When compared with k-means which only has an average value of Silhouette Score 0.6872873866.

scholarly journals Identifying Patterns of Students Academic Performance from Tracer Evaluation using Descriptive Data Mining

2019 ◽  
Vol 8 (2S2) ◽  
pp. 187-191
Keyword(s):  
Data Mining ◽  
Academic Performance ◽  
English Language ◽  
Public Universities ◽  
Study Data ◽  
Performance Model ◽  
Tracer Study ◽  
Descriptive Data ◽  
Hidden Knowledge ◽  
The Impact

The Ministry of Higher Education Malaysia has collected data through tracer study since 2007. The aim is to gather feedbacks from graduates as a basis improve to basis in improving. The availability of tracer study data in digital format offers various advantages to decision makers as many tools are available to extract and discover the hidden knowledge within the large databases. This paper presents the applicability of descriptive data mining and logistic regression to discover the hidden knowledge within the tracer study data with respect to measuring academic performance of Arts and Sciences graduates of Malaysia public universities. The impact of independent variables, i.e. Bahasa Melayu, English Language and Malaysian University English Test on the academic performance is investigated. The empirical results suggest that the academic performance between male and female graduates from Arts and Science fields is significantly different. Variables such as Bahasa Melayu, English Language and Malaysian University English Test showed a significant correlation with academic performance. The results also exhibit that the impact on academic performance of Arts graduates is different from the Science graduates. Guided by these empirical findings, this study suggests an academic performance model for Arts and Science graduates of Malaysia public universities.

Data Mining and Machine Learning

2020 ◽  
Author(s):  
Mohammed J. Zaki ◽  
Wagner Meira, Jr
Keyword(s):  
Machine Learning ◽  
Data Mining

Heart Rate Variability, Emotions, and Music

2010 ◽  
Vol 24 (2) ◽  
pp. 112-119 ◽  
Author(s):  
F. Riganello ◽  
A. Candelieri ◽  
M. Quintieri ◽  
G. Dolce
Keyword(s):  
Data Mining ◽  
Heart Rate ◽  
Heart Rate Variability ◽  
Vegetative State ◽  
Low Frequency ◽  
Emotional Reactions ◽  
Heart Beat ◽  
Healthy Controls ◽  
Frequency Spectra ◽  
Emotional Value

The purpose of the study was to identify significant changes in heart rate variability (an emerging descriptor of emotional conditions; HRV) concomitant to complex auditory stimuli with emotional value (music). In healthy controls, traumatic brain injured (TBI) patients, and subjects in the vegetative state (VS) the heart beat was continuously recorded while the subjects were passively listening to each of four music samples of different authorship. The heart rate (parametric and nonparametric) frequency spectra were computed and the spectra descriptors were processed by data-mining procedures. Data-mining sorted the nu_lf (normalized parameter unit of the spectrum low frequency range) as the significant descriptor by which the healthy controls, TBI patients, and VS subjects’ HRV responses to music could be clustered in classes matching those defined by the controls and TBI patients’ subjective reports. These findings promote the potential for HRV to reflect complex emotional stimuli and suggest that residual emotional reactions continue to occur in VS. HRV descriptors and data-mining appear applicable in brain function research in the absence of consciousness.

Sign in / Sign up

Export Citation Format

Share Document