diff --git a/content/en/projects/biome-classification.md b/content/en/projects/biome-classification.md
index 092102f..a247472 100644
--- a/content/en/projects/biome-classification.md
+++ b/content/en/projects/biome-classification.md
@@ -4,14 +4,12 @@ title = "Random forest models for the prediction of biome types and climate vari
date = "2024-11-14"
+++
-{{< notice tip >}} The paper is available [here](/files/random-forests-biomes.pdf). {{< /notice >}}
+{{< notice tip >}} The report is available [here](/files/random-forests-biomes.pdf). {{< /notice >}}
# Reflection
This project was a big learning moment when it comes selecting training and testing datasets appropriately in statistical learning. The model can only ever be as good as the data we use. It was one of the first times working with geographical, grid-based data, which was also interesting. Since all of the data worked with was directly fed from a model, it's also important to know the limits of one's original model which provided the data. Sometimes, the problem may not be our classifier or regression model, but simply that we did not have enough, or the right, information to properly distinguish data in the first place.
-This project strengthened my skills in building basic random forest pipelines, from data partitioning and preprocessing to hyperparameter tuning, performance evaluation, and model interpretation, all within the context of environmental and climate data.
-
# Summary
In this project, I developed some random forest models to predict biome classes (both binary and multi-class) as well as two continuous climate variables: vegetation carbon pool (VegC) and net primary productivity (NPP). To adress class imbalances, I used SMOTE up-sampling, which notably improved recall for underrepresented classes. By grid-search cross-validation I tried to tune the models better. Performance was evaluated using accuracy, precision, recall, and F₁ score for classifiers, and RMSE for regressors. Finally, I analyzed feature importance to better understand which climate variables, such as seasonal precipitation or extreme temperatures, were driving the model predictions.
diff --git a/public/en/index.xml b/public/en/index.xml
index d65ae0e..2e4a537 100644
--- a/public/en/index.xml
+++ b/public/en/index.xml
@@ -1,47 +1,98 @@
-
-
-
- Pim Nelissen
- //localhost:1313/en/
- Recent content on Pim Nelissen
- Hugo
- en
- Thu, 27 Jun 2024 00:00:00 +0000
-
-
- Machine learning applied to radioactive decay data (Bachelor thesis)
- //localhost:1313/en/projects/bsc-thesis/
- Thu, 27 Jun 2024 00:00:00 +0000
- //localhost:1313/en/projects/bsc-thesis/
- <div class="notice tip">
<div class="notice-title">
<i class="fa-solid fa-lightbulb" aria-hidden="true"></i>Tip
</div>
<div class="notice-content">The paper is available for <a href="http://lup.lub.lu.se/student-papers/record/9168893" class="external-link" target="_blank" rel="noopener">open access</a>.</div>
</div>
<div class="notice tip">
<div class="notice-title">
<i class="fa-solid fa-lightbulb" aria-hidden="true"></i>Tip
</div>
<div class="notice-content">The code I wrote for the decay chain simulations is openly available on my <a href="https://git.pimnel.com/pim/shn-decay-chains" class="external-link" target="_blank" rel="noopener">Git instance</a>.</div>
</div>
<h1 id="reflection">
Reflection
<a class="heading-link" href="#reflection">
<i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
<span class="sr-only">Link to heading</span>
</a>
</h1>
<p>This 6 month journey has been one with many challenges, but also one of a lot of personal growth. Life was not always easy during this semester, which had caused unfortunate delays in the project. As my supervisor assured me, this is how it always goes in research. Plans continuously change, timetables need adjusting and ambitious ideas put on hold. This project then certainly made me improve my planning skills. Dealing with setbacks was also more common in this rather independent project, compared to any coursework before.</p>
-
-
- Coupled Pendula & The Kuramoto Model (1st year experimental project)
- //localhost:1313/en/projects/first-year-project/
- Tue, 15 Jun 2021 00:00:00 +0000
- //localhost:1313/en/projects/first-year-project/
- <div class="notice tip">
<div class="notice-title">
<i class="fa-solid fa-lightbulb" aria-hidden="true"></i>Tip
</div>
<div class="notice-content"><p>Lund University wrote an article about our project. You can find it <a href="https://www.fysik.lu.se/artikel/pandemisakert-nar-fysikstudenter-redovisar" class="external-link" target="_blank" rel="noopener">here</a>.</p>
<p>(Note: the article is written in Swedish).</p></div>
</div>
<div style="position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;">
<iframe allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen="allowfullscreen" loading="eager" referrerpolicy="strict-origin-when-cross-origin" src="https://www.youtube.com/embed/roK2gRDHWeE?autoplay=0&controls=1&end=0&loop=0&mute=0&start=0" style="position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;" title="YouTube video"></iframe>
</div>
-
-
- Contact
- //localhost:1313/en/contact/
- Mon, 01 Jan 0001 00:00:00 +0000
- //localhost:1313/en/contact/
- <p>Email: <a href="mailto:pi0274ne-s@student.lu.se" >pi0274ne-s@student.lu.se</a></p>
-
-
- CV
- //localhost:1313/en/cv/
- Mon, 01 Jan 0001 00:00:00 +0000
- //localhost:1313/en/cv/
- <head>
<link rel="stylesheet" href="//localhost:1313/css/timeline.css">
</head>
<h2 id="-education">
🎓 Education
<a class="heading-link" href="#-education">
<i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
<span class="sr-only">Link to heading</span>
</a>
</h2>
<div class="timeline">
<div class="timeline-item">
<div class="dot"></div>
<div class="timeline-date">2024 - <i>now</i></div>
<div class="timeline-content">
<h4>M.Sc. in Computational Science (Physics)</h4>
<p><i class="fas fa-university"></i> Lund University, Sweden 🇸🇪</p>
<p><code>Numerical ODE/PDE solvers</code> <code>MATLAB</code> <code>Monte Carlo</code> <code>Subatomic Physics</code></p>
</div>
</div>
<div class="timeline-item">
<div class="dot"></div>
<div class="timeline-date">2020 - 2024</div>
<div class="timeline-content">
<h4>B.Sc. in Physics</h4>
<p><i class="fas fa-university"></i> Lund University, Sweden 🇸🇪</p>
<p><code>Python</code> <code>Deep learning</code> <code>Monte Carlo</code> <code>C#</code> <code>SQL</code></p>
<li class="timeline-content">Bachelor thesis: <i>"Scrutinizing the Schmidt Test and Exploring the Use of Machine Learning for Statistical Assessment of Radioactive Decay Chains"</i>, Lund University Publications, 2024. <a href="http://lup.lub.lu.se/student-papers/record/9168893" target="_blank">Open Access</a></li>
</div>
</div>
</div>
<hr>
<h2 id="-professional-experience">
💼 Professional Experience
<a class="heading-link" href="#-professional-experience">
<i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
<span class="sr-only">Link to heading</span>
</a>
</h2>
<div class="timeline">
<div class="timeline-item">
<div class="dot"></div>
<div class="timeline-date">2018 - 2020</div>
<div class="timeline-content">
<h4>Sales Specialist</h4>
<p><i class="fas fa-briefcase"></i> Jumbo Supermarkten, Netherlands 🇳🇱</p>
-
-
- Skills
- //localhost:1313/en/skills/
- Mon, 01 Jan 0001 00:00:00 +0000
- //localhost:1313/en/skills/
- <h2 id="core-competencies">
Core Competencies
<a class="heading-link" href="#core-competencies">
<i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
<span class="sr-only">Link to heading</span>
</a>
</h2>
<p>From my background in physics and computational science, I have a strong foundation in numerical methods and Monte Carlo simulations for solving physical problems. I am particularly interested by nuclear physics. My <a href="http://lup.lub.lu.se/student-papers/record/9168893" class="external-link" target="_blank" rel="noopener">bachelor degree thesis</a> is about applying machine learning on radioactive decay data. I am proficient in Python and have some exposure to C#, MATLAB, and SQL. I am very comfortable with GNU/Linux and Git workflows.</p>
-
-
-
+Pim Nelissen/en/Recent content on Pim NelissenHugoenThu, 14 Nov 2024 00:00:00 +0000Random forest models for the prediction of biome types and climate variables/en/projects/biome-classification/Thu, 14 Nov 2024 00:00:00 +0000/en/projects/biome-classification/<div class="notice tip">
+ <div class="notice-title">
+ <i class="fa-solid fa-lightbulb" aria-hidden="true"></i>Tip
+ </div>
+ <div class="notice-content">The paper is available <a href="/files/random-forests-biomes.pdf" >here</a>.</div>
+</div>
+
+<h1 id="reflection">
+ Reflection
+ <a class="heading-link" href="#reflection">
+ <i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
+ <span class="sr-only">Link to heading</span>
+ </a>
+</h1>
+<p>This project was a big learning moment when it comes selecting training and testing datasets appropriately in statistical learning. The model can only ever be as good as the data we use. It was one of the first times working with geographical, grid-based data, which was also interesting. Since all of the data worked with was directly fed from a model, it’s also important to know the limits of one’s original model which provided the data. Sometimes, the problem may not be our classifier or regression model, but simply that we did not have enough, or the right, information to properly distinguish data in the first place.</p>Machine learning applied to radioactive decay data (Bachelor thesis)/en/projects/bsc-thesis/Thu, 27 Jun 2024 00:00:00 +0000/en/projects/bsc-thesis/<div class="notice tip">
+ <div class="notice-title">
+ <i class="fa-solid fa-lightbulb" aria-hidden="true"></i>Tip
+ </div>
+ <div class="notice-content">The paper is available for <a href="http://lup.lub.lu.se/student-papers/record/9168893" class="external-link" target="_blank" rel="noopener">open access</a>.</div>
+</div>
+
+<div class="notice tip">
+ <div class="notice-title">
+ <i class="fa-solid fa-lightbulb" aria-hidden="true"></i>Tip
+ </div>
+ <div class="notice-content">The code I wrote for the decay chain simulations is openly available on my <a href="https://git.pimnel.com/pim/shn-decay-chains" class="external-link" target="_blank" rel="noopener">Git instance</a>.</div>
+</div>
+
+<h1 id="reflection">
+ Reflection
+ <a class="heading-link" href="#reflection">
+ <i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
+ <span class="sr-only">Link to heading</span>
+ </a>
+</h1>
+<p>This 6 month journey has been one with many challenges, but also one of a lot of personal growth. Life was not always easy during this semester, which had caused unfortunate delays in the project. As my supervisor assured me, this is how it always goes in research. Plans continuously change, timetables need adjusting and ambitious ideas put on hold. This project then certainly made me improve my planning skills. Dealing with setbacks was also more common in this rather independent project, compared to any coursework before.</p>Coupled Pendula & The Kuramoto Model (1st year experimental project)/en/projects/first-year-project/Tue, 15 Jun 2021 00:00:00 +0000/en/projects/first-year-project/<div class="notice tip">
+ <div class="notice-title">
+ <i class="fa-solid fa-lightbulb" aria-hidden="true"></i>Tip
+ </div>
+ <div class="notice-content"><p>Lund University wrote an article about our project. You can find it <a href="https://www.fysik.lu.se/artikel/pandemisakert-nar-fysikstudenter-redovisar" class="external-link" target="_blank" rel="noopener">here</a>.</p>
+<p>(Note: the article is written in Swedish).</p></div>
+</div>
+
+<div style="position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden;">
+ <iframe allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" allowfullscreen="allowfullscreen" loading="eager" referrerpolicy="strict-origin-when-cross-origin" src="https://www.youtube.com/embed/roK2gRDHWeE?autoplay=0&controls=1&end=0&loop=0&mute=0&start=0" style="position: absolute; top: 0; left: 0; width: 100%; height: 100%; border:0;" title="YouTube video"></iframe>
+ </div>Contact/en/contact/Mon, 01 Jan 0001 00:00:00 +0000/en/contact/<p>Email: <a href="mailto:pi0274ne-s@student.lu.se" >pi0274ne-s@student.lu.se</a></p>CV/en/cv/Mon, 01 Jan 0001 00:00:00 +0000/en/cv/<head>
+ <link rel="stylesheet" href="/css/timeline.css">
+</head>
+<h2 id="-education">
+ 🎓 Education
+ <a class="heading-link" href="#-education">
+ <i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
+ <span class="sr-only">Link to heading</span>
+ </a>
+</h2>
+<div class="timeline">
+ <div class="timeline-item">
+ <div class="dot"></div>
+ <div class="timeline-date">2024 - <i>now</i></div>
+ <div class="timeline-content">
+ <h4>M.Sc. in Computational Science (Physics)</h4>
+ <p><i class="fas fa-university"></i> Lund University, Sweden 🇸🇪</p>
+ <p><code>Numerical ODE/PDE solvers</code> <code>MATLAB</code> <code>Monte Carlo</code> <code>Subatomic Physics</code></p>
+ </div>
+ </div>
+ <div class="timeline-item">
+ <div class="dot"></div>
+ <div class="timeline-date">2020 - 2024</div>
+ <div class="timeline-content">
+ <h4>B.Sc. in Physics</h4>
+ <p><i class="fas fa-university"></i> Lund University, Sweden 🇸🇪</p>
+ <p><code>Python</code> <code>Deep learning</code> <code>Monte Carlo</code> <code>C#</code> <code>SQL</code></p>
+ <li class="timeline-content">Bachelor thesis: <i>"Scrutinizing the Schmidt Test and Exploring the Use of Machine Learning for Statistical Assessment of Radioactive Decay Chains"</i>, Lund University Publications, 2024. <a href="http://lup.lub.lu.se/student-papers/record/9168893" target="_blank">Open Access</a></li>
+ </div>
+ </div>
+</div>
+<hr>
+<h2 id="-professional-experience">
+ 💼 Professional Experience
+ <a class="heading-link" href="#-professional-experience">
+ <i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
+ <span class="sr-only">Link to heading</span>
+ </a>
+</h2>
+<div class="timeline">
+ <div class="timeline-item">
+ <div class="dot"></div>
+ <div class="timeline-date">2018 - 2020</div>
+ <div class="timeline-content">
+ <h4>Sales Specialist</h4>
+ <p><i class="fas fa-briefcase"></i> Jumbo Supermarkten, Netherlands 🇳🇱</p>Skills/en/skills/Mon, 01 Jan 0001 00:00:00 +0000/en/skills/<h2 id="core-competencies">
+ Core Competencies
+ <a class="heading-link" href="#core-competencies">
+ <i class="fa-solid fa-link" aria-hidden="true" title="Link to heading"></i>
+ <span class="sr-only">Link to heading</span>
+ </a>
+</h2>
+<p>From my background in physics and computational science, I have a strong foundation in numerical methods and Monte Carlo simulations for solving physical problems. I am particularly interested by nuclear physics. My <a href="http://lup.lub.lu.se/student-papers/record/9168893" class="external-link" target="_blank" rel="noopener">bachelor degree thesis</a> is about applying machine learning on radioactive decay data. I am proficient in Python and have some exposure to C#, MATLAB, and SQL. I am very comfortable with GNU/Linux and Git workflows.</p>
\ No newline at end of file
diff --git a/public/en/projects/biome-classification/index.html b/public/en/projects/biome-classification/index.html
new file mode 100644
index 0000000..d5dfa91
--- /dev/null
+++ b/public/en/projects/biome-classification/index.html
@@ -0,0 +1,29 @@
+Random forest models for the prediction of biome types and climate variables · Pim Nelissen
+
This project was a big learning moment when it comes selecting training and testing datasets appropriately in statistical learning. The model can only ever be as good as the data we use. It was one of the first times working with geographical, grid-based data, which was also interesting. Since all of the data worked with was directly fed from a model, it’s also important to know the limits of one’s original model which provided the data. Sometimes, the problem may not be our classifier or regression model, but simply that we did not have enough, or the right, information to properly distinguish data in the first place.
This project strengthened my skills in building basic random forest pipelines, from data partitioning and preprocessing to hyperparameter tuning, performance evaluation, and model interpretation, all within the context of environmental and climate data.
In this project, I developed some random forest models to predict biome classes (both binary and multi-class) as well as two continuous climate variables: vegetation carbon pool (VegC) and net primary productivity (NPP). To adress class imbalances, I used SMOTE up-sampling, which notably improved recall for underrepresented classes. By grid-search cross-validation I tried to tune the models better. Performance was evaluated using accuracy, precision, recall, and F₁ score for classifiers, and RMSE for regressors. Finally, I analyzed feature importance to better understand which climate variables, such as seasonal precipitation or extreme temperatures, were driving the model predictions.
The binary biome classifier achieved up to 85.7% accuracy after SMOTE resampling, effectively distinguishing between temperate deciduous and mixed forests, with winter temperature and autumn precipitation emerging as key predictors. The multi-class classifier reached a weighted F₁ score of around 0.65, although it struggled to separate closely related biomes, reflecting the continuity between the biomes proves challenging for ML to solve. The regression models performed well overall, but revealed spatial biases around coastal and desert areas, suggesting the need to account for additional local processes like soil variability or ocean influences.
This 6 month journey has been one with many challenges, but also one of a lot of personal growth. Life was not always easy during this semester, which had caused unfortunate delays in the project. As my supervisor assured me, this is how it always goes in research. Plans continuously change, timetables need adjusting and ambitious ideas put on hold. This project then certainly made me improve my planning skills. Dealing with setbacks was also more common in this rather independent project, compared to any coursework before.
+
On an intellectual level, it was a challenge in the beginning to adjust to the academic writing style of papers. Information is more dense and sometimes not explained in detail, referring the reader to some other paper. This trail of research papers that one can go down, even in a small topic as the one I was researching, is truly eye-opening to me. The amount of information that is out there is unimaginable to me. The project also introduced me to new applications of statistics and tested some old statistical knowledge, such as that on exponential distributions, statistical moments like variance and higher order ones. On top of that, I enjoyed learning about Monte Carlo simulations. I had always figured this term to sound complicated, but to learn that it is a rather easy methodology which has such strong use cases in various real-world applications was very interesting to learn. While not the main focus of my thesis, I also enjoyed researching the experimental background, of the process of superheavy synthesis and how detection is done.
Experimental nuclear structure data coming from superheavy nuclei synthesis experiments often consists of correlated alpha decay chains. In the absence of neutron detectors - which would fully characterize the exit channel after the fusion-evaporation reaction - the sequence of decay energies and half-lives are the ‘fingerprint’ of the exit channel itself. Experimental data in this region is sparse, and its interpretation can be liable to error or confirmation bias. A so-called “Schmidt test” is a method for determining the congruence of correlation times for a set of measurements of one decay step. Its outcome is not always entirely conclusive, however. This study evaluates the congruence derived from the Schmidt test using Monte Carlo simulated data with various level of contamination from incongruent data. Furthermore, the study also includes the evaluation of congruence of data stemming from single decays and multi-step decay chains. A multi-layer perceptron was trained on extracted features from simulated decay chain sets with one step. The Schmidt test performs well with larger decay sets and when the half-life of the contaminating species is longer than the original species by a factor 5 or 10. However, the test performs poorly in low counting statistics, where few recorded decay times are available. The newly proposed machine learning model outperforms the Schmidt test in certain high statistics scenarios, but also fails when few decay times are available. Its performance is also poor when the half-life of the contaminant is shorter than the original half-life. The learning behaviour of the model is analysed, showing significant contributions from higher statistical moments in training. Future work involves including chain correlations across multiple steps, alpha decay energies, as well as the potential use of alternative machine learning models.
One of the key results of the thesis was the heatmap in Figure 1, showing how the amount of contamination in a decay time data set (that is, number of decay times from a species other than the one we are interested in) affects the Schmidt test and its conclusions. This was tested for various factors of contaminant half lives. The result is that the Schmidt test seems rather unsensitive to any contaminants with a half life on the same order of magnitude as the species we seek to study. This was further motivation to look beyond the Schmidt test, which solely relies on variance of the dataset, to see if other statistical methods (e.g. machine learning) could be used.
+
+
Figure 1: Heatmaps for the Schmidt test congruence for various contaminant half-lives. Each pixel is a unique combination of set size N and C contaminated decay times, where the value of the pixel represents the percentage of i=100 simulated decay chain sets with j=1 step for which the measure σ(Θ,exp) falls within the confidence interval, indicating likely congruence. The longer the contaminant half life is compared to the original, the more sets are deemed incongruent by the Schmidt test. Additionally, it can be seen that when nearly all of the decay times in the data set are from the contaminant species, the data becomes congruent again. The number of sets deemed congruent is somewhat symmetrical around the the line of 50% contamination.
This 6 month journey has been one with many challenges, but also one of a lot of personal growth. Life was not always easy during this semester, which had caused unfortunate delays in the project. As my supervisor assured me, this is how it always goes in research. Plans continuously change, timetables need adjusting and ambitious ideas put on hold. This project then certainly made me improve my planning skills. Dealing with setbacks was also more common in this rather independent project, compared to any coursework before.
-
On an intellectual level, it was a challenge in the beginning to adjust to the academic writing style of papers. Information is more dense and sometimes not explained in detail, referring the reader to some other paper. This trail of research papers that one can go down, even in a small topic as the one I was researching, is truly eye-opening to me. The amount of information that is out there is unimaginable to me. The project also introduced me to new applications of statistics and tested some old statistical knowledge, such as that on exponential distributions, statistical moments like variance and higher order ones. On top of that, I enjoyed learning about Monte Carlo simulations. I had always figured this term to sound complicated, but to learn that it is a rather easy methodology which has such strong use cases in various real-world applications was very interesting to learn. While not the main focus of my thesis, I also enjoyed researching the experimental background, of the process of superheavy synthesis and how detection is done.
Experimental nuclear structure data coming from superheavy nuclei synthesis experiments often consists of correlated alpha decay chains. In the absence of neutron detectors - which would fully characterize the exit channel after the fusion-evaporation reaction - the sequence of decay energies and half-lives are the ‘fingerprint’ of the exit channel itself. Experimental data in this region is sparse, and its interpretation can be liable to error or confirmation bias. A so-called “Schmidt test” is a method for determining the congruence of correlation times for a set of measurements of one decay step. Its outcome is not always entirely conclusive, however. This study evaluates the congruence derived from the Schmidt test using Monte Carlo simulated data with various level of contamination from incongruent data. Furthermore, the study also includes the evaluation of congruence of data stemming from single decays and multi-step decay chains. A multi-layer perceptron was trained on extracted features from simulated decay chain sets with one step. The Schmidt test performs well with larger decay sets and when the half-life of the contaminating species is longer than the original species by a factor 5 or 10. However, the test performs poorly in low counting statistics, where few recorded decay times are available. The newly proposed machine learning model outperforms the Schmidt test in certain high statistics scenarios, but also fails when few decay times are available. Its performance is also poor when the half-life of the contaminant is shorter than the original half-life. The learning behaviour of the model is analysed, showing significant contributions from higher statistical moments in training. Future work involves including chain correlations across multiple steps, alpha decay energies, as well as the potential use of alternative machine learning models.
One of the key results of the thesis was the heatmap in Figure 1, showing how the amount of contamination in a decay time data set (that is, number of decay times from a species other than the one we are interested in) affects the Schmidt test and its conclusions. This was tested for various factors of contaminant half lives. The result is that the Schmidt test seems rather unsensitive to any contaminants with a half life on the same order of magnitude as the species we seek to study. This was further motivation to look beyond the Schmidt test, which solely relies on variance of the dataset, to see if other statistical methods (e.g. machine learning) could be used.
-
-
Figure 1: Heatmaps for the Schmidt test congruence for various contaminant half-lives. Each pixel is a unique combination of set size N and C contaminated decay times, where the value of the pixel represents the percentage of i=100 simulated decay chain sets with j=1 step for which the measure σ(Θ,exp) falls within the confidence interval, indicating likely congruence. The longer the contaminant half life is compared to the original, the more sets are deemed incongruent by the Schmidt test. Additionally, it can be seen that when nearly all of the decay times in the data set are from the contaminant species, the data becomes congruent again. The number of sets deemed congruent is somewhat symmetrical around the the line of 50% contamination.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
+This 6 month journey has been one with many challenges, but also one of a lot of personal growth. Life was not always easy during this semester, which had caused unfortunate delays in the project. As my supervisor assured me, this is how it always goes in research. Plans continuously change, timetables need adjusting and ambitious ideas put on hold. This project then certainly made me improve my planning skills. Dealing with setbacks was also more common in this rather independent project, compared to any coursework before.">
This 6 month journey has been one with many challenges, but also one of a lot of personal growth. Life was not always easy during this semester, which had caused unfortunate delays in the project. As my supervisor assured me, this is how it always goes in research. Plans continuously change, timetables need adjusting and ambitious ideas put on hold. This project then certainly made me improve my planning skills. Dealing with setbacks was also more common in this rather independent project, compared to any coursework before.
On an intellectual level, it was a challenge in the beginning to adjust to the academic writing style of papers. Information is more dense and sometimes not explained in detail, referring the reader to some other paper. This trail of research papers that one can go down, even in a small topic as the one I was researching, is truly eye-opening to me. The amount of information that is out there is unimaginable to me. The project also introduced me to new applications of statistics and tested some old statistical knowledge, such as that on exponential distributions, statistical moments like variance and higher order ones. On top of that, I enjoyed learning about Monte Carlo simulations. I had always figured this term to sound complicated, but to learn that it is a rather easy methodology which has such strong use cases in various real-world applications was very interesting to learn. While not the main focus of my thesis, I also enjoyed researching the experimental background, of the process of superheavy synthesis and how detection is done.
Experimental nuclear structure data coming from superheavy nuclei synthesis experiments often consists of correlated alpha decay chains. In the absence of neutron detectors - which would fully characterize the exit channel after the fusion-evaporation reaction - the sequence of decay energies and half-lives are the ‘fingerprint’ of the exit channel itself. Experimental data in this region is sparse, and its interpretation can be liable to error or confirmation bias. A so-called “Schmidt test” is a method for determining the congruence of correlation times for a set of measurements of one decay step. Its outcome is not always entirely conclusive, however. This study evaluates the congruence derived from the Schmidt test using Monte Carlo simulated data with various level of contamination from incongruent data. Furthermore, the study also includes the evaluation of congruence of data stemming from single decays and multi-step decay chains. A multi-layer perceptron was trained on extracted features from simulated decay chain sets with one step. The Schmidt test performs well with larger decay sets and when the half-life of the contaminating species is longer than the original species by a factor 5 or 10. However, the test performs poorly in low counting statistics, where few recorded decay times are available. The newly proposed machine learning model outperforms the Schmidt test in certain high statistics scenarios, but also fails when few decay times are available. Its performance is also poor when the half-life of the contaminant is shorter than the original half-life. The learning behaviour of the model is analysed, showing significant contributions from higher statistical moments in training. Future work involves including chain correlations across multiple steps, alpha decay energies, as well as the potential use of alternative machine learning models.
One of the key results of the thesis was the heatmap in Figure 1, showing how the amount of contamination in a decay time data set (that is, number of decay times from a species other than the one we are interested in) affects the Schmidt test and its conclusions. This was tested for various factors of contaminant half lives. The result is that the Schmidt test seems rather unsensitive to any contaminants with a half life on the same order of magnitude as the species we seek to study. This was further motivation to look beyond the Schmidt test, which solely relies on variance of the dataset, to see if other statistical methods (e.g. machine learning) could be used.
Figure 1: Heatmaps for the Schmidt test congruence for various contaminant half-lives. Each pixel is a unique combination of set size N and C contaminated decay times, where the value of the pixel represents the percentage of i=100 simulated decay chain sets with j=1 step for which the measure σ(Θ,exp) falls within the confidence interval, indicating likely congruence. The longer the contaminant half life is compared to the original, the more sets are deemed incongruent by the Schmidt test. Additionally, it can be seen that when nearly all of the decay times in the data set are from the contaminant species, the data becomes congruent again. The number of sets deemed congruent is somewhat symmetrical around the the line of 50% contamination.
