nuluh/thesis
1
0
Fork 0
Code Issues 47 Pull Requests Actions Packages Projects Releases 1 Wiki Activity

feat(latex): add glossary, acronyms, and notations

Browse Source
This commit is contained in:
nuluh
2025-06-04 14:36:00 +07:00
parent cec43cb291
commit 1914cc3bf7
5 changed files with 346 additions and 63 deletions
Download Patch File Download Diff File Expand all files Collapse all files

12
latex/bibliography.bib
View File

@@ -963,3 +963,15 @@
@thesis{zotero-622,
type = {thesis}
}
@thesis{rytter1993,
title = {Vibrational {{Based Inspection}} of {{Civil Engineering Structures}}},
author = {Rytter, Anders},
date = {1993},
institution = {Aalborg University},
location = {Aalborg},
url = {https://vbn.aau.dk/en/publications/vibrational-based-inspection-of-civil-engineering-structures},
abstract = {The thesis has been written in relation to two different research projects. Firstly, an offshore test programme, Integrated Experimental/Numerical Analysis of the Dynamic behavior of offshore structures, which was performed at the department of Building Technology and Structural Engineering at the University of Aalborg from 1988 to 1991. Secondly, a research project, In-Field Vibration Based Inspection of Civil Engineering Structures, which has been performed as a pilot project by the Consulting Engineers Rambøll, Hannemann and Højlund in cooperation with the department of Building Technology and Structural Engineering at the University of Aalborg since the beginning of 1992. Both projects have been supported by the Danish Technical Research Council. Further, the first mentioned project was supported by the Danish Energy Agency. Their financial support is gratefully acknowledged.},
langid = {english},
keywords = {Beam,Bridges,Cracks,Damping,Offshore Platform,Piles,Structural Damage,VBI,Vibration Based Inspection}
}

15
latex/frontmatter/acronym.tex Normal file
View File

@@ -0,0 +1,15 @@
% Define an abbreviation (acronym)
% Acronyms for the thesis
\newacronym{ml}{ML}{machine learning}
\newacronym{stft}{STFT}{short-time fourier transform}
\newacronym{ai}{AI}{artificial intelligence}
\newacronym{dl}{DL}{deep learning}
\newacronym{nn}{NN}{neural network}
\newacronym{fft}{FFT}{fast fourier transform}
\newacronym{svm}{SVM}{support vector machine}
\newacronym{cnn}{CNN}{convolutional neural network}
\newacronym{rnn}{RNN}{recurrent neural network}
\newacronym{vbi}{VBI}{vibration-based inspection}
\newacronym{shm}{SHM}{structural health monitoring}
\newacronym{fea}{FEA}{finite element analysis}
\newacronym{1d-cnn}{1-D CNN}{\textit{One-Dimensional Convolutional Neural Network}}

132
latex/frontmatter/glossaries.tex
View File

@@ -1,46 +1,3 @@
% % A new command that enables us to enter bi-lingual (Slovene and English) terms
% % syntax: \addterm[options]{label}{Slovene}{Slovene first use}{English}{Slovene
% % description}
% \newcommand{\addterm}[6][]{
% \newglossaryentry{#2}{
% name={#3 (angl.\ #5)},
% first={#4 (\emph{#5})},
% text={#3},
% sort={#3},
% description={#6},
% #1 % pass additional options to \newglossaryentry
% }
% }
% % A new command that enables us to enter (English) acronyms with bi-lingual
% % (Slovene and English) long versions
% % syntax: \addacronym[options]{label}{abbreviation}{Slovene long}{Slovene first
% % use long}{English long}{Slovene description}
% \newcommand{\addacronym}[7][]{
% % Create the main glossary entry with \newacronym
% % \newacronym[key-val list]{label}{abbrv}{long}
% \newacronym[
% name={#4 (angl.\ #6,\ #3)},
% first={\emph{#5} (angl.\ \emph{#6},\ \emph{#3})},
% sort={#4},
% description={#7},
% #1 % pass additional options to \newglossaryentry
% ]
% {#2}{#3}{#4}
% % Create a cross-reference from the abbreviation to the main glossary entry by
% % creating an auxiliary glossary entry (note: we set the label of this entry
% % to '<original label>_auxiliary' to avoid clashes)
% \newglossaryentry{#2_auxiliary}{
% name={#3},
% sort={#3},
% description={\makefirstuc{#6}},
% see=[See:]{#2}
% }
% }
% % Change the text of the cross-reference links to the Slovene long version.
% \renewcommand*{\glsseeitemformat}[1]{\emph{\acrlong{#1}}.}
% Define the Indonesian term and link it to the English term
\newglossaryentry{jaringansaraf}{
name=Jaringan Saraf,
@@ -52,27 +9,78 @@
% }
% Define the English term and link it to its acronym
\newglossaryentry{neuralnetwork}{
name=Neural Network,
description={A computational model inspired by the human brain, see \gls{nn}}
}
% \newglossaryentry{neuralnetwork}{
% name=Neural Network,
% description={A computational model inspired by the human brain, see \gls{nn}}
% }
% \newacronym
% [description={statistical pattern recognition technique}]
% {svm}{SVM}{support vector machine}
% \newglossaryentry{machinelearning}{
% name=Machine Learning,
% description={A program or system that trains a model from input data. The trained model can make useful predictions from new (never-before-seen) data drawn from the same distribution as the one used to train the model.}}
% \newglossaryentry{pemelajaranmesin}{
% name={pemelajaran mesin (angl.\ #5)},
% first={pemelajaran mesin (\emph{machine learning})},
% text={pemelajaran mesin},
% sort={ },
% description={#6},
% #1 % pass additional options to \newglossaryentry
% }
\longnewglossaryentry{machinelearning}{name={machine learning}}
{A program or system that trains a model from input data. The trained model can make useful predictions from new (never-before-seen) data drawn from the same distribution as the one used to train the model.}
\newterm[see={machinelearning}]{pemelajaranmesin}
% \longnewglossaryentry{machinelearning}{name={machine learning}}
% {A program or system that trains a model from input data. The trained model can make useful predictions from new (never-before-seen) data drawn from the same distribution as the one used to train the model.}
% \newterm[see={machinelearning}]{pemelajaranmesin}
% \newglossaryentry{pemelajaran mesin}{}
% \addterm{machinelearning}{pemelajaran mesin}{pemelajaran mesin}{machine learning}{A program or system that trains a model from input data. The trained model can make useful predictions from new (never-before-seen) data drawn from the same distribution as the one used to train the model.}
\newacronym
[description={statistical pattern recognition technique}]
{svm}{SVM}{support vector machine}
\newglossaryentry{algoritma-genetika}{
name={Algoritma Genetika},
description={Kelas algoritma optimasi dan pencarian yang terinspirasi oleh proses evolusi biologis, seperti seleksi alam, mutasi, dan rekombinasi. Algoritma ini sering digunakan untuk menemukan solusi perkiraan untuk masalah yang kompleks dan sulit dipecahkan secara analitis.},
sort={Algoritma Genetika}
}
\newglossaryentry{deep-learning}{
name={\textit{deep learning}},
description={Bagian dari keluarga metode pembelajaran mesin yang lebih luas berdasarkan jaringan saraf tiruan dengan banyak lapisan (deep neural networks). Arsitektur ini memungkinkan model untuk belajar representasi data secara hierarkis, mulai dari fitur tingkat rendah hingga konsep abstrak tingkat tinggi.},
sort={Pembelajaran Mendalam}
}
\newglossaryentry{jaringan-saraf-tiruan}{
name={Jaringan Saraf Tiruan (Artificial Neural Network)},
description={Model komputasi yang terinspirasi oleh struktur dan fungsi jaringan saraf biologis di otak. JST terdiri dari unit pemrosesan yang saling terhubung (neuron) yang bekerja secara paralel untuk memproses informasi dan belajar dari data melalui penyesuaian bobot koneksi.},
sort={Jaringan Saraf Tiruan}
}
\newglossaryentry{pemrosesan-bahasa-alami}{
name={Pemrosesan Bahasa Alami (Natural Language Processing)},
description={Cabang ilmu komputer dan kecerdasan buatan yang berfokus pada interaksi antara komputer dan bahasa manusia. Tujuannya adalah untuk memungkinkan komputer memproses, memahami, menafsirkan, dan menghasilkan bahasa manusia dengan cara yang bermakna dan berguna.},
sort={Pemrosesan Bahasa Alami}
}
\newglossaryentry{pembelajaran-penguatan}{
name={Pembelajaran Penguatan (Reinforcement Learning)},
description={Area pembelajaran mesin yang berkaitan dengan bagaimana agen perangkat lunak harus mengambil tindakan dalam suatu lingkungan untuk memaksimalkan beberapa gagasan tentang imbalan kumulatif. Agen belajar melalui trial-and-error, menerima umpan balik berupa imbalan atau hukuman.},
sort={Pembelajaran Penguatan}
}
\newglossaryentry{visi-komputer}{
name={Visi Komputer (Computer Vision)},
description={Bidang interdisipliner yang membahas bagaimana komputer dapat dibuat untuk mendapatkan pemahaman tingkat tinggi dari gambar atau video digital. Dari perspektif rekayasa, ia berupaya mengotomatiskan tugas-tugas yang dapat dilakukan oleh sistem visual manusia.},
sort={Visi Komputer}
}
\newglossaryentry{model-generatif}{
name={Model Generatif},
description={Jenis model statistik dalam pembelajaran mesin yang bertujuan untuk mempelajari distribusi probabilitas dari data pelatihan. Setelah dilatih, model ini dapat menghasilkan sampel data baru yang mirip dengan data pelatihan, seperti membuat gambar, teks, atau suara baru.},
sort={Model Generatif}
}
\newglossaryentry{heuristik}{
name={Heuristik},
description={Teknik pemecahan masalah yang menggunakan pendekatan praktis atau jalan pintas yang tidak dijamin optimal atau sempurna, tetapi cukup untuk mencapai tujuan jangka pendek atau perkiraan solusi. Heuristik sering digunakan ketika pencarian solusi optimal terlalu mahal secara komputasi.},
sort={Heuristik}
}
\newglossaryentry{validasi-silang}{
name={Validasi Silang (Cross-Validation)},
description={Teknik statistik untuk mengevaluasi seberapa baik hasil analisis statistik (seperti model prediktif) akan generalisasi ke kumpulan data independen. Ini penting untuk menghindari overfitting dan mendapatkan estimasi kinerja model yang lebih andal pada data yang belum pernah dilihat.},
sort={Validasi Silang}
}
\newglossaryentry{bias-algoritmik}{
name={Bias Algoritmik},
description={Mengacu pada kesalahan sistematis atau hasil yang tidak adil yang dihasilkan oleh sistem kecerdasan buatan karena asumsi yang salah dalam proses pembelajaran mesin atau karena data pelatihan yang bias. Bias ini dapat mereplikasi atau bahkan memperkuat prasangka sosial yang ada.},
sort={Bias Algoritmik}
}

241
latex/frontmatter/notations.tex Normal file
View File

@@ -0,0 +1,241 @@
% --- Glossary Definitions ---
% Note: Descriptions are based on the provided Indonesian text but translated to English
% for typical glossary conventions. You can adjust the language as needed.
\newglossaryentry{not:signal}{
name={\ensuremath{S}},
description={vektor sinyal akselerometer berdimensi 1$\times$262144},
sort={s},
type=notation,
}
\newglossaryentry{not:sampling_freq}{
name={\ensuremath{f_s}},
description={frekuensi dengan nilai \textit{sampling} ($s$) di mana sinyal kontinu didigitalkan},
sort={fs},
type=notation,
}
\newglossaryentry{not:time_length}{
name={\ensuremath{t}},
description={panjang waktu data dalam detik},
sort={t},
type=notation,
}
\newglossaryentry{not:dataset_A}{
name={\ensuremath{\mathcal{A}}},
description={matriks dataset A},
sort={adataset},
type=notation,
}
\newglossaryentry{not:dataset_B}{
name={\ensuremath{\mathcal{B}}},
description={matriks dataset B},
sort={bdataset},
type=notation,
}
\newglossaryentry{not:damage_file}{
name={\ensuremath{\mathbf{D}}},
description={matriks akselerometer untuk setiap berkas dengan bentuk $262144\times30$},
sort={filedamage},
type=notation,
}
\newglossaryentry{not:joint_index}{
name={\ensuremath{n}},
description={indeks atau nomor kerusakan \textit{joint}},
sort={indexjoint},
type=notation,
}
\newglossaryentry{not:damage_file_set_case}{
name={\ensuremath{\mathbf{d}}},
description={set matriks kerusakan},
sort={damagefilesetcase},
type=notation,
}
\newglossaryentry{not:k}{
name={$k$},
description={Index for measurement nodes, an integer ranging from 0 to 29.},
sort={k},
type=notation,
}
\newglossaryentry{not:Fk}{
name={$F_{k}$},
description={Filename string for the raw time-domain signal from node $k$. The specific format mentioned is \texttt{zzzAD}$k$\texttt{.TXT}.},
sort={Fk},
type=notation,
}
\newglossaryentry{not:nkFk}{
name={$n_{k}^{F_{k}}$},
description={Represents the measurement \textit{node} with index $k$. The raw time-domain signal data from this node, $x_k$, has a length of $L=262144$ samples.},
sort={nkFk},
type=notation,
}
\newglossaryentry{not:i}{
name={$i$},
description={Index for ``damage-case'' folders, an integer ranging from 0 to 5.},
sort={i},
type=notation,
}
\newglossaryentry{not:di}{
name={$d_{i}$},
description={Set representing the $i$-th damage scenario, containing data from five consecutive nodes: $\bigl\{\,n_{5i}^{F_{5i}},\;n_{5i+1}^{F_{5i+1}},\;\dots,\;n_{5i+4}^{F_{5i+4}}\bigr\}$. Cardinality: $|d_i|=5$ nodes.},
sort={di},
type=notation,
}
\newglossaryentry{not:diTD}{
name={$d_{i}^{\mathrm{TD}}$},
description={Time-domain subset of nodes from damage case $d_i$, containing only the first and last nodes: $\bigl\{\,n_{5i}^{F_{5i}},\;n_{5i+4}^{F_{5i+4}}\bigr\}$. Cardinality: $|d_{i}^{\mathrm{TD}}| = 2$ nodes.},
sort={diTD},
type=notation,
}
\newglossaryentry{not:calT}{
name={$\mathcal{T}$},
description={Short-Time Fourier Transform (STFT) operator. It maps a raw time-domain signal $n_k^{F_k}$ (or $x_k$) from $\mathbb{R}^{L}$ (with $L=262144$) to a magnitude spectrogram matrix $\widetilde{n}_k^{F_k}$ in $\mathbb{R}^{513 \times 513}$.},
sort={Tcal},
type=notation,
}
\newglossaryentry{not:L}{
name={$L$},
description={Length of the raw time-domain signal, $L=262144$ samples.},
sort={L},
type=notation,
}
\newglossaryentry{not:Nw}{
name={$N_{w}$},
description={Length of the Hanning window used in the STFT, $N_{w}=1024$ samples.},
sort={Nw},
type=notation,
}
\newglossaryentry{not:Nh}{
name={$N_{h}$},
description={Hop size (or step size) used in the STFT, $N_{h}=512$ samples.},
sort={Nh},
type=notation,
}
\newglossaryentry{not:wn}{
name={$w[n]$},
description={Value of the Hanning window function at sample index $n$. The window spans $N_w$ samples.},
sort={wn},
type=notation,
}
\newglossaryentry{not:n_summation}{
name={$n$},
description={Sample index within the Hanning window and for the STFT summation, an integer ranging from $0$ to $N_w-1$.},
sort={n_summation},
type=notation,
}
\newglossaryentry{not:xkm}{
name={$x_k[m]$}, % Or x_k if it's treated as the whole signal vector
description={Represents the raw time-domain signal for node $k$. As a discrete signal, it consists of $L=262144$ samples. $x_k[m]$ would be the $m$-th sample.},
sort={xkm},
type=notation,
}
\newglossaryentry{not:Skpt}{
name={$S_k(p,t)$},
description={Complex-valued result of the STFT for node $k$ at frequency bin $p$ and time frame $t$. This is a scalar value for each $(p,t)$ pair.},
sort={Skpt},
type=notation,
}
\newglossaryentry{not:p}{
name={$p$},
description={Frequency bin index in the STFT or spectrogram, an integer ranging from $0$ to $512$.},
sort={p},
type=notation,
}
\newglossaryentry{not:t_stft}{ % Differentiating t for STFT time frame and t for feature vector time slice if necessary
name={$t$},
description={Time frame index in the STFT or spectrogram, an integer ranging from $0$ to $512$. Also used as the time slice index for extracting feature vectors $\mathbf{x}_{i,s,r,t}$ from spectrograms.},
sort={t},
type=notation,
}
\newglossaryentry{not:ntildekFk}{ % New entry for the matrix
name={$\widetilde{n}_k^{F_k}$},
description={The magnitude spectrogram matrix for node $k$, obtained by applying the STFT operator $\mathcal{T}$ to the time-domain signal $n_k^{F_k}$. This matrix is an element of $\mathbb{R}^{513 \times 513}$.},
sort={ntildekFk},
type=notation,
}
\newglossaryentry{not:ntildekFkpt}{ % Modified entry for the element
name={$\widetilde{n}_k^{F_k}(p,t)$},
description={Scalar value representing the magnitude of the STFT for node $k$ at frequency bin $p$ and time frame $t$; specifically, $\widetilde{n}_k^{F_k}(p,t) = |S_k(p,t)|$. This is an element of the spectrogram matrix $\widetilde{n}_k^{F_k}$.},
sort={ntildekFkpt},
type=notation,
}
\newglossaryentry{not:R}{
name={$\mathbb{R}$},
description={The set of real numbers. Used to denote vector spaces like $\mathbb{R}^{N}$ (N-dimensional real vectors) or $\mathbb{R}^{M \times N}$ (M-by-N real matrices).},
sort={Rbb},
type=notation,
}
\newglossaryentry{not:diFD}{
name={$d_{i}^{\mathrm{FD}}$},
description={Frequency-domain subset for damage case $i$. It contains two spectrogram matrices: $\bigl\{\,\widetilde{n}_{5i}^{F_{5i}},\; \widetilde{n}_{5i+4}^{F_{5i+4}}\,\bigr\}$, where each spectrogram $\widetilde{n}$ is in $\mathbb{R}^{513 \times 513}$. Cardinality: $|d_{i}^{\mathrm{FD}}| = 2$ spectrograms.},
sort={diFD},
type=notation,
}
\newglossaryentry{not:r_repetition}{
name={$r$},
description={Repetition index within a single damage case, an integer ranging from $0$ to $4$.},
sort={r_repetition},
type=notation,
}
\newglossaryentry{not:xboldisr}{
name={$\mathbf{x}_{i,s,r,t}$},
description={Feature vector (a row or column, often referred to as a time slice) taken from the $r$-th spectrogram repetition, for damage case $i$ and sensor side $s$, at time slice $t$. This vector is an element of $\mathbb{R}^{513}$.},
sort={xisrt_bold},
type=notation,
}
\newglossaryentry{not:s_sensor}{
name={$s$},
description={Index representing the sensor side (e.g., identifying Sensor A or Sensor B).},
sort={s_sensor},
type=notation,
}
\newglossaryentry{not:yi}{
name={$y_{i}$},
description={Scalar label for the damage case $i$, defined as $y_i = i$. This is an integer value from 0 to 5.},
sort={yi},
type=notation,
}
\newglossaryentry{not:Lambda}{
name={$\Lambda(i,s,r,t)$},
description={Slicing function that concatenates a feature vector $\mathbf{x}_{i,s,r,t} \in \mathbb{R}^{513}$ with its corresponding damage case label $y_i \in \mathbb{R}$, resulting in a combined vector $\bigl[\,\mathbf{x}_{i,s,r,t}, \;y_{i}\bigr] \in \mathbb{R}^{514}$.},
sort={Lambda},
type=notation,
}
\newglossaryentry{not:calDs}{
name={$\mathcal{D}^{(s)}$},
description={The complete dataset for sensor side $s$. It is a collection of $15390$ data points, where each point is a vector in $\mathbb{R}^{514}$ (513 features + 1 label). Thus, the dataset can be viewed as a matrix of size $15390 \times 514$.},
sort={Dcal_s},
type=notation,
}
% --- End Glossary Definitions ---

7
latex/main.tex
View File

@@ -19,6 +19,13 @@
% \input{preamble/fonts}
\input{preamble/macros}
\addbibresource{bibliography.bib}
\makeglossaries
\input{frontmatter/acronym}
\input{frontmatter/glossaries}
\input{frontmatter/notations}
\begin{document}
% \input{frontmatter/maketitle}
% \input{frontmatter/maketitle_secondary}