@@ -35,12 +35,10 @@ <h1>My Projects</h1>
3535 < div class ="container ">
3636 < div class ="filter-buttons ">
3737 < button class ="filter-btn active " data-filter ="all "> All Projects</ button >
38- < button class ="filter-btn " data-filter ="web "> Web Apps</ button >
39- < button class ="filter-btn " data-filter ="mobile "> Mobile Apps</ button >
40- < button class ="filter-btn " data-filter ="api "> APIs</ button >
41- < button class ="filter-btn " data-filter ="tools "> Tools</ button >
38+ < button class ="filter-btn " data-filter ="web "> AI Applications</ button >
4239 < button class ="filter-btn " data-filter ="robotics "> Robotics</ button >
4340 < button class ="filter-btn " data-filter ="ml "> Machine Learning</ button >
41+ < button class ="filter-btn " data-filter ="tools "> Research & Analysis</ button >
4442 </ div >
4543 </ div >
4644 </ section >
@@ -51,7 +49,7 @@ <h1>My Projects</h1>
5149
5250 < div class ="project-card featured " data-category ="web ">
5351 < div class ="project-image ">
54- < img src ="images/language-tutor-app .jpg " alt ="Neo4j Language Learning Tutor ">
52+ < img src ="images/neo4j .jpg " alt ="Neo4j Language Learning Tutor ">
5553 < div class ="project-overlay ">
5654 < div class ="project-links ">
5755 < a href ="https://github.com/coding-crying/realtime-agents-language-tutor " class ="btn btn-primary " target ="_blank "> GitHub</ a >
@@ -90,45 +88,6 @@ <h4>Key Features:</h4>
9088 </ div >
9189 </ div >
9290
93- < div class ="project-card " data-category ="ml ">
94- < div class ="project-image ">
95- < img src ="images/ml-model-comparison.png " alt ="Movie Score Prediction ML Model ">
96- < div class ="project-overlay ">
97- < div class ="project-links ">
98- < a href ="https://github.com/MirkoRUG/Applied-ML-Group17 " class ="btn btn-primary " target ="_blank "> GitHub</ a >
99- </ div >
100- </ div >
101- </ div >
102- < div class ="project-info ">
103- < h3 > Movie Score Prediction ML System</ h3 >
104- < p class ="project-description ">
105- End-to-end machine learning pipeline using CatBoost regression to predict IMDb
106- movie scores. Ensemble models learning achieved excellent performance with
107- RMSE ~0.3, demonstrating comprehensive hyperparameter optimization and production deployment.
108- </ p >
109- < div class ="tech-stack ">
110- < span class ="tech-tag "> CatBoost</ span >
111- < span class ="tech-tag "> Optuna</ span >
112- < span class ="tech-tag "> FastAPI</ span >
113- < span class ="tech-tag "> Streamlit</ span >
114- < span class ="tech-tag "> Docker</ span >
115- < span class ="tech-tag "> Python</ span >
116- </ div >
117- < div class ="project-features ">
118- < h4 > Key Achievements:</ h4 >
119- < ul >
120- < li > Excellent ensemble model performance (RMSE ~0.3)</ li >
121- < li > 17% performance improvement over baseline methods</ li >
122- < li > Bayesian hyperparameter optimization with 50 trials</ li >
123- < li > Production-ready FastAPI deployment with Docker</ li >
124- < li > Comprehensive evaluation with 5-fold cross-validation</ li >
125- < li > Interactive Streamlit web interface for predictions</ li >
126- < li > Robust preprocessing pipeline with outlier detection</ li >
127- </ ul >
128- </ div >
129- </ div >
130- </ div >
131-
13291 < div class ="project-card " data-category ="robotics ">
13392 < div class ="project-image ">
13493 < img src ="images/hoody-robot.jpg " alt ="Hoody - Smart Kitchen Hood Robot ">
@@ -169,6 +128,45 @@ <h4>Research Contributions:</h4>
169128 </ div >
170129 </ div >
171130
131+ < div class ="project-card " data-category ="ml ">
132+ < div class ="project-image ">
133+ < img src ="images/ml-model-comparison.png " alt ="Movie Score Prediction ML Model ">
134+ < div class ="project-overlay ">
135+ < div class ="project-links ">
136+ < a href ="https://github.com/MirkoRUG/Applied-ML-Group17 " class ="btn btn-primary " target ="_blank "> GitHub</ a >
137+ </ div >
138+ </ div >
139+ </ div >
140+ < div class ="project-info ">
141+ < h3 > Movie Score Prediction ML System</ h3 >
142+ < p class ="project-description ">
143+ End-to-end machine learning pipeline using CatBoost regression to predict IMDb
144+ movie scores. Ensemble models learning achieved excellent performance with
145+ RMSE ~0.3, demonstrating comprehensive hyperparameter optimization and production deployment.
146+ </ p >
147+ < div class ="tech-stack ">
148+ < span class ="tech-tag "> CatBoost</ span >
149+ < span class ="tech-tag "> Optuna</ span >
150+ < span class ="tech-tag "> FastAPI</ span >
151+ < span class ="tech-tag "> Streamlit</ span >
152+ < span class ="tech-tag "> Docker</ span >
153+ < span class ="tech-tag "> Python</ span >
154+ </ div >
155+ < div class ="project-features ">
156+ < h4 > Key Achievements:</ h4 >
157+ < ul >
158+ < li > Excellent ensemble model performance (RMSE ~0.3)</ li >
159+ < li > 17% performance improvement over baseline methods</ li >
160+ < li > Bayesian hyperparameter optimization with 50 trials</ li >
161+ < li > Production-ready FastAPI deployment with Docker</ li >
162+ < li > Comprehensive evaluation with 5-fold cross-validation</ li >
163+ < li > Interactive Streamlit web interface for predictions</ li >
164+ < li > Robust preprocessing pipeline with outlier detection</ li >
165+ </ ul >
166+ </ div >
167+ </ div >
168+ </ div >
169+
172170 < div class ="project-card " data-category ="tools ">
173171 < div class ="project-image ">
174172 < img src ="images/patent-design.jpg " alt ="Patented Braking System Design ">
@@ -210,6 +208,37 @@ <h4>Development Achievements:</h4>
210208 </ div >
211209
212210
211+ < div class ="project-card " data-category ="ml ">
212+ < div class ="project-info ">
213+ < h3 > Cognitive Modeling of Mind-Wandering and Rumination</ h3 >
214+ < p class ="project-description ">
215+ Research investigating spreading activation in emotional memory and its effects on
216+ mind-wandering episodes using ACT-R cognitive architecture. Explores how memory
217+ activation capacity influences rumination persistence and attention task performance.
218+ </ p >
219+ < div class ="tech-stack ">
220+ < span class ="tech-tag "> ACT-R</ span >
221+ < span class ="tech-tag "> Cognitive Modeling</ span >
222+ < span class ="tech-tag "> SART</ span >
223+ < span class ="tech-tag "> Memory Activation</ span >
224+ < span class ="tech-tag "> Statistical Analysis</ span >
225+ </ div >
226+ < div class ="project-features ">
227+ < h4 > Research Findings:</ h4 >
228+ < ul >
229+ < li > Computational model of rumination using spreading activation</ li >
230+ < li > ANOVA analysis showing significant differences across conditions</ li >
231+ < li > Evidence that higher activation capacity leads to persistent rumination</ li >
232+ < li > Sustained Attention to Response Task (SART) simulation</ li >
233+ < li > Quantitative support for negative memory bias theories</ li >
234+ </ ul >
235+ </ div >
236+ < div class ="project-links ">
237+ < a href ="Mind Wandering Cognitive Modelling.pdf " class ="btn btn-primary " target ="_blank "> Download Paper</ a >
238+ </ div >
239+ </ div >
240+ </ div >
241+
213242 < div class ="project-card " data-category ="tools ">
214243 < div class ="project-image ">
215244 < img src ="images/domain-analysis.jpg " alt ="TrueSelect AI Recruitment Analysis ">
@@ -246,37 +275,6 @@ <h4>Key Contributions:</h4>
246275 </ div >
247276 </ div >
248277
249- < div class ="project-card " data-category ="ml ">
250- < div class ="project-info ">
251- < h3 > Cognitive Modeling of Mind-Wandering and Rumination</ h3 >
252- < p class ="project-description ">
253- Research investigating spreading activation in emotional memory and its effects on
254- mind-wandering episodes using ACT-R cognitive architecture. Explores how memory
255- activation capacity influences rumination persistence and attention task performance.
256- </ p >
257- < div class ="tech-stack ">
258- < span class ="tech-tag "> ACT-R</ span >
259- < span class ="tech-tag "> Cognitive Modeling</ span >
260- < span class ="tech-tag "> SART</ span >
261- < span class ="tech-tag "> Memory Activation</ span >
262- < span class ="tech-tag "> Statistical Analysis</ span >
263- </ div >
264- < div class ="project-features ">
265- < h4 > Research Findings:</ h4 >
266- < ul >
267- < li > Computational model of rumination using spreading activation</ li >
268- < li > ANOVA analysis showing significant differences across conditions</ li >
269- < li > Evidence that higher activation capacity leads to persistent rumination</ li >
270- < li > Sustained Attention to Response Task (SART) simulation</ li >
271- < li > Quantitative support for negative memory bias theories</ li >
272- </ ul >
273- </ div >
274- < div class ="project-links ">
275- < a href ="Mind Wandering Cognitive Modelling.pdf " class ="btn btn-primary " target ="_blank "> Download Paper</ a >
276- </ div >
277- </ div >
278- </ div >
279-
280278 </ div >
281279 </ div >
282280 </ section >
0 commit comments