Garuda - Garba Rujukan Digital

IDENTIFIKASI SPESIES MANGROVE MENGGUNAKAN ALGORITME RANDOM FOREST Ishak Ariawan; Ayang Armelita Rosalia; Luthfi Anzani; Wildan Aprizal Arifin; La Ode Alam Minsaris; Lukman Lukman
Jurnal Kemaritiman: Indonesian Journal of Maritime Vol 2, No 2 (2021): Desember 2021
Publisher : Universitas Pendidikan Indonesia (UPI)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Identifikasi komposisi spesies mangrove adalah topik yang penting dalam manajemen dan konservasi ekosistem pesisir. Biodiversitas spesies mangrove berpengaruh terhadap keberlangsungan dan keseimbangan entitas-entitas yang terkait di dalam ekosistemnya. Penelitian ini bertujuan untuk menggali potensi Machine Learning untuk mengidentifikasi spesies mangrove. Secara spesifik, algortime Random Forest digunakan untuk mengklasifikasikan enam spesies mangrove, yaitu: Avicennia eucalyptifolia, Bruguiera gymnorrhiza, Rhizophora apiculata, Rhizophora mucronata, Unrecorded Sp., dan Xylocarpus granatum. Beberapa pendekatan dilakukan untuk memperkuat kinerja algoritme Random Forest, yaitu preprocessing (SMOTE) dan normalisasi min-max untuk menyeimbangkan distribusi data. Hasil tahap ini menunjukkan bahwa proyeksi dari rentang normalisasi (interval 0-1) tidak berpengaruh dalam mereduksi pola data secara dimensional. Setelah preprocessing dan normalisasi data, lima atribut (spesies, wood density, diameter at beast height, total of above ground biomass, dan below-ground root) diklasifikasi dan dianalisis dengan spesies sebagai atribut target. Pembangunan parameter model didasarkan pada jumlah total dari hasil SMOTE dengan menetapkan 100 dan 500 sebagai jumlah pohon tunggal dan 1000 sebagai jumlah node dan peubah prediktor default. Hasil akhir menunjukkan bahwa algortime Random Forest memperoleh nilai evaluasi yang optimal dengan rata-rata 99.97% menggunakan jumlah pohon tunggal dan cut-off yang telah ditetapkan. Akurasi maksimal yaitu 100% diperoleh dari jumlah pohon tunggal dan cut-off dengan ukuran sebagai berikut: (1) 500 dan 80:20; (2) 500 dan 90:10; dan (3) 100 dan 80:20. Hasil ini menunjukkan bahwa algortime Random Forest sangat efektif untuk diterapkan sebagai metode pengklasifikasi Machine Learning dalam menentukan spesies mangrove.

Karakteristik estimator Analisis Komponen Utama untuk mengestimasi Model Variabel Laten menggunakan metode High-Dimensional AIC Lukman Lukman
Jurnal EurekaMatika Vol 9, No 1 (2021): Jurnal Eurekamatika
Publisher : Universitas Pendidikan Indonesia (UPI)

Makalah ini bertujuan untuk mengetahui sifat estimator Analisis Komponen Utama (AKU) untuk mengestimasi model variabel laten. Metode yang digunakan adalah metode High-Dimensional AIC (HAIC) dengan simulasi data berdistribusi Bernoulli. Tahapannya adalah: (1) menentukan matriks AKU; (2) membuat model estimator AKU untuk mengestimasi variabel laten dengan menggunakan HAIC; (3) mensimulasikan data distribusi Bernoulli dengan pengulangan 1.000.748 kali. Hasil simulasi menunjukkan model estimator AKU bekerja dengan baik.

STUDYING STUDENT STATISTICAL LITERACY IN STATISTICS LECTURES ON HIGHER EDUCATION USING GROUNDED THEORY APPROACH Lukman Lukman; Wahyudin Wahyudin; Didi Suryadi; Dadan Dasari; Sufyani Prabawanto
Jurnal Infinity Vol 11, No 1 (2022): VOLUME 11, NUMBER 1, INFINITY
Publisher : IKIP Siliwangi and I-MES

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22460/infinity.v11i1.p163-176

The purpose of this study was to obtain an overview of the student's Statistical Literacy model in Statistics learning in Higher Education. Researchers conducted an in-depth study of student statistical literacy, how they understand and apply statistics, how statistics are used as a tool for reliable data that can be trusted as scientific works. The research method uses qualitative research methods with a Grounded Theory approach. Participants involved in this study were 114 participants from several universities in West Java, Indonesia. The results of this study found a student statistical literacy model consisting of 2 dimensions and 5 elements. Dimensions of Statistical Knowledge: Descriptive Statistics, Inference Statistics, Statistical Communication and Statistical Reasoning. Attitude Dimensions: Confidence and Critical Attitude.

Mekanisme Pengumpulan Data Hasil Tangkapan di Pelabuhan Perikanan Nusantara (PPN) Karangantu Ayang Armelita Rosalia; Luthfi Anzhani; La Ode Alam Minsaris; Denta Tirtana; Lukman Lukman; Abdul Malik; Cakra Rahardjo
Baselang Vol 4, No 1: APRIL 2024
Publisher : Fakultas Pertanian Universitas Muara Bungo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36355/bsl.v4i1.135

Pendataan hasil tangkapan merupakan salah satu aktivitas untuk mengetahui jumlah dan jenis ikan hasil tangkapan dari suatu kapal perikanan. Kegiatan ini dilakukan untuk mendapatkan data yang realistik terkait hasil tangkapan yang didaratkan sehingga data yang didapatkan dapat digunakan oleh Pelabuhan Perikanan sebagai pembanding logbook yang diserahkan oleh pihak kapal, sehingga didapatkan data produktivitas yang akurat. Optimalisasi kinerja petugas pendataan akan meningkatkan efektivitas, struktur dan tugas organisasi pelaksana pendataan. Berdasarkan hal tersebut maka dalam penelitian ini akan menganalisis bagaimana kinerja sistem pendataan hasil tangkapan ikan yang didaratkan di PPN Karangantu dan merekomendasikan saran tindakan perbaikan sistem pendataan hasil tangkapan ikan di PPN Karangantu. Hasil penelitian yang didapatkan adalah pada tahap analisis sudah dilakukan wawancara dengan nelayan, enumerator dan syahbandar PPN Karangantu. Standar Operasional Prosedur (SOP) yang digunakan di PPN Karangantu adalah sudah sesuai dengan SOP yang berlaku secara nasional yaitu Direktorat Jenderal Perikanan Tangkap, Direktorat Pelabuhan Perikanan, dengan pengadopsian dan modifikasi SOP sehingga terjadi penyesuaian SOP yang diberlakukan di PPN Karangantu. Ditemukan beberapa kekurangan pada sistem pendataan yang sudah diterapkan di PPN Karangantu beberapa di antaranya yaitu terbatasnya Sumber Daya Manusia (SDM) untuk mendata ke lokasi-lokasi pendataan hasil tangkapan yang didaratkan di luar Tempat Pendaratan Ikan (TPI). Potensi atau saran untuk efisiensi pendataan hasil tangkap di PPN Karangantu adalah adanya aplikasi yang dapat mengidentifikasi jenis ikan untuk entry pendataan yang diikuti hasil penimbangan berkaitan dengan efisiensi waktu sehingga dapat mengurangi delay.

STUDYING STUDENT STATISTICAL LITERACY IN STATISTICS LECTURES ON HIGHER EDUCATION USING GROUNDED THEORY APPROACH Lukman Lukman; Wahyudin Wahyudin; Didi Suryadi; Dadan Dasari; Sufyani Prabawanto
Jurnal Infinity Vol 11 No 1 (2022): VOLUME 11, NUMBER 1, INFINITY
Publisher : IKIP Siliwangi and I-MES

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.22460/infinity.v11i1.p163-176

The purpose of this study was to obtain an overview of the student's Statistical Literacy model in Statistics learning in Higher Education. Researchers conducted an in-depth study of student statistical literacy, how they understand and apply statistics, how statistics are used as a tool for reliable data that can be trusted as scientific works. The research method uses qualitative research methods with a Grounded Theory approach. Participants involved in this study were 114 participants from several universities in West Java, Indonesia. The results of this study found a student statistical literacy model consisting of 2 dimensions and 5 elements. Dimensions of Statistical Knowledge: Descriptive Statistics, Inference Statistics, Statistical Communication and Statistical Reasoning. Attitude Dimensions: Confidence and Critical Attitude.

A Study Of Differences In Mathematical Reasoning Among Junior High Students With Field Independent And Field Dependent Cognitive Styles Fujiama Marjud; Aan Hasanah; Lukman Lukman; Putri Ardhanita Harahap
Mathline : Jurnal Matematika dan Pendidikan Matematika Vol. 10 No. 4 (2025): Mathline : Jurnal Matematika dan Pendidikan Matematika
Publisher : Universitas Wiralodra

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31943/mathline.v10i4.1049

Mathematical reasoning plays a vital role in helping students understand concepts, identify patterns, and solve problems effectively. This study aimed to investigate whether students with Field-independent (FI) and Field-dependent (FD) cognitive styles exhibit significant differences in their mathematical reasoning abilities. To examine the variation in students’ reasoning skills according to their cognitive styles, a quantitative approach using a comparative method was employed. After a preliminary test of homogeneity of variances using Levene's Test, which produced a significance value of 0.820 (> 0.05), Data were analyzed using an Independent Samples t-test. The results indicated that the assumption of equal variances was met, thereby validating the use of parametric analysis. The t-test revealed a statistically significant difference between the two cognitive style groups, with a t-value of 3.232, 40 degrees of freedom, and a significance level of 0.002 (< 0.05). In terms of mathematical reasoning, students identified as FI performed better than those classified as FD. This finding suggests that while FD learners tend to rely more on external cues and structured guidance when solving problems, FI learners are generally more analytical and able to process information independently. The findings also suggest that pupils' capacity for mathematical reasoning is significantly influenced by their cognitive style. These results emphasise the necessity of varied teaching strategies that take into account both learning styles from a pedagogical standpoint. Teachers can build more inclusive, balanced, and productive mathematics learning environments by incorporating exploratory, Problem-based tasks are more suitable for FI learners, whereas structured instructional approaches better support FD learners.

Analisis Learning Obstacle Siswa pada Materi Peluang Berbasis Literasi Matematis dengan Pendekatan Didactical Design Research Rachmawati Khaerun Nisa; Lukman Lukman; Sufyani Prabawanto
MUST: Journal of Mathematics Education, Science and Technology Vol 10 No 2 (2025): DECEMBER
Publisher : Universitas Muhammadiyah Surabaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30651/must.v10i2.32981

Mathematical literacy is a fundamental competency for the twenty-first century; however, students continue to experience significant learning obstacles in probability. This study aims to conduct an in-depth analysis of the learning obstacles encountered by eleventh-grade senior high school students in probability within the context of PISA-based mathematical literacy using the Didactical Design Research (DDR) approach. The study was conducted at a senior high school in Kuningan Regency, West Java, Indonesia, involving 36 students who participated in the diagnostic test stage. Using purposive sampling based on the variation of learning obstacle typologies (ontogenic, didactical, and epistemological), 10 students were selected for in-depth interviews. The data were analyzed qualitatively through data triangulation, including data reduction, data display, and conclusion drawing. The findings revealed that, at the formulate stage, students experienced epistemological and ontogenic obstacles characterized by difficulties in abstracting verbal situations into formal mathematical models and by mathematical anxiety. At the employ stage, substantial epistemological obstacles emerged due to inadequate prerequisite knowledge of set theory, procedural errors in fraction operations, and excessive reliance on memorizing formulas. Meanwhile, at the interpret stage, students were unable to relate numerical results back to their real-world contextual meanings. The novelty of this study lies in its simultaneous mapping of learning obstacles across the three PISA mathematical literacy processes—formulate, employ, and interpret—which served as the foundation for designing Didactical Pedagogical Anticipation (DPA) and reconstructing the recommended Hypothetical Learning Trajectory (HLT) from three to six instructional sessions.

Title

Found 7 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 7 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 7 Documents
Search