Desain dan Analisis Algoritma Komputasi Matriks pada Penyelesaian Regresi Multilinier: Studi Kasus Beecrowd 1819 - Estimating Production

Hanin, Shafa Nabilah (2025) Desain dan Analisis Algoritma Komputasi Matriks pada Penyelesaian Regresi Multilinier: Studi Kasus Beecrowd 1819 - Estimating Production. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025211222-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (5MB) | Request a copy

Abstract

Regresi multilinier adalah model regresi dengan satu variabel dependen dan lebih dari satu variabel independen. Model ini cocok diterapkan pada kasus yang membutuhkan parameter seminimal mungkin namun tetap bermakna secara kontekstual. Tugas akhir ini mengkaji penerapan regresi multilinier berbasis komputasi matriks pada studi kasus “Beecrowd 1819 – Estimating Production”, yang bertujuan memprediksi hasil produksi pertanian berdasarkan dua variabel input: luas lahan dan jumlah tenaga kerja. Dalam pendekatan ini, regresi multilinier digunakan untuk meminimalkan kesalahan kuadrat antara prediksi dan data aktual dengan menghitung koefisien. Metode yang digunakan adalah least squares, dengan pendekatan komputasi matriks untuk menghitung parameter model secara efisien. Implementasi dilakukan dalam bahasa pemrograman C++ menggunakan kelas matriks buatan sendiri. Pengujian dilakukan untuk mengevaluasi kebenaran solusi dan efisiensi waktu eksekusi. Hasil uji coba menunjukkan bahwa algoritma dapat menghasilkan output dengan rata-rata waktu eksekusi 0,01 detik dan memori 5,18 MB dalam sepuluh kali pengujian. Hal ini menunjukkan bahwa pendekatan regresi multilinier berbasis komputasi matriks mampu menghasilkan solusi yang akurat dan efisien.
==================================================================================================================================
Multilinear regression is a regression model that involves one dependent variable and more than one independent variable. This model is well-suited for cases that require a minimal number of parameters while maintaining contextual relevance. This thesis explores the application of matrix-based multilinear regression through the case study "Beecrowd 1819 – Estimating Production", which aims to predict agricultural production outcomes based on two input variables: land area and labor force. In this approach, multilinear regression is employed to minimize the squared error between predicted and actual values by calculating appropriate coefficients. The method used is least squares, implemented using a matrix computation technique to efficiently determine the model parameters. The implementation is carried out in C++, utilizing a custom-built matrix class. Testing is performed to assess both the correctness of the solution and the efficiency of its execution time. The results indicate that the algorithm produces accurate output with an average execution time of 0.01 seconds and memory usage of 5.18 MB across ten trials. These results demonstrate that the matrix-based multilinear regression approach provides both accuracy and computational efficiency.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	komputasi matriks, least squares, regresi multilinier, least squares, matrix computation, multilinear regression
Subjects:	Q Science > QA Mathematics > QA184 Algebra, Linear Q Science > QA Mathematics > QA275 Theory of errors. Least squares. Including statistical inference. Error analysis (Mathematics) Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA9.58 Algorithms
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Shafa Nabilah Hanin
Date Deposited:	10 Jul 2025 04:02
Last Modified:	10 Jul 2025 04:02
URI:	http://repository.its.ac.id/id/eprint/119480

Actions (login required)

View Item