| | SLO | ENG | Piškotki in zasebnost

Večja pisava | Manjša pisava

Iskanje po katalogu digitalne knjižnice Pomoč

Iskalni niz: išči po
išči po
išči po
išči po
* po starem in bolonjskem študiju

Opcije:
  Ponastavi


1 - 10 / 55
Na začetekNa prejšnjo stran123456Na naslednjo stranNa konec
1.
FairBoost: Boosting supervised learning for learning on multiple sensitive features
Ivona Colakovic, Sašo Karakatič, 2023, izvirni znanstveni članek

Opis: The vast majority of machine learning research focuses on improving the correctness of the outcomes (i.e., accuracy, error-rate, and other metrics). However, the negative impact of machine learning outcomes can be substantial if the consequences marginalize certain groups of data, especially if certain groups of people are the ones being discriminated against. Thus, recent papers try to tackle the unfair treatment of certain groups of data (humans), but mostly focus on only one sensitive feature with binary values. In this paper, we propose an ensemble boosting FairBoost that takes into consideration fairness as well as accuracy to mitigate unfairness in classification tasks during the model training process. This method tries to close the gap between proposed approaches and real-world applications, where there is often more than one sensitive feature that contains multiple categories. The proposed approach checks the bias and corrects it through the iteration of building the boosted ensemble. The proposed FairBoost is tested within the experimental setting and compared to similar existing algorithms. The results on different datasets and settings show no significant changes in the overall quality of classification, while the fairness of the outcomes is vastly improved.
Ključne besede: fairness, boosting, machine learning, supervised learning
Objavljeno v DKUM: 11.06.2024; Ogledov: 62; Prenosov: 2
.pdf Celotno besedilo (1,70 MB)
Gradivo ima več datotek! Več...

2.
Adaptive boosting method for mitigating ethnicity and age group unfairness
Ivona Colakovic, Sašo Karakatič, 2024, izvirni znanstveni članek

Opis: Machine learning algorithms make decisions in various fields, thus influencing people’s lives. However, despite their good quality, they can be unfair to certain demographic groups, perpetuating socially induced biases. Therefore, this paper deals with a common unfairness problem, unequal quality of service, that appears in classification when age and ethnicity groups are used. To tackle this issue, we propose an adaptive boosting algorithm that aims to mitigate the existing unfairness in data. The proposed method is based on the AdaBoost algorithm but incorporates fairness in the calculation of the instance’s weight with the goal of making the prediction as good as possible for all ages and ethnicities. The results show that the proposed method increases the fairness of age and ethnicity groups while maintaining good overall quality compared to traditional classification algorithms. The proposed method achieves the best accuracy in almost every sensitive feature group. Based on the extensive analysis of the results, we found that when it comes to ethnicity, interestingly, White people are likely to be incorrectly classified as not being heroin users, whereas other groups are likely to be incorrectly classified as heroin users.
Ključne besede: fairness, boosting, machine learning, classification
Objavljeno v DKUM: 24.05.2024; Ogledov: 231; Prenosov: 7
.pdf Celotno besedilo (1,66 MB)
Gradivo ima več datotek! Več...

3.
Attachment in close relationships and glycemic outcomes in children with type 1 diabetes
Simona Klemenčič, Jasna Klara Lipovšek, Anja Turin, Klemen Dovč, Nataša Bratina, Yael Shmueli-Goetz, Katarina Trebušak Podkrajšek, Barbka Repič-Lampret, Barbara Jenko Bizjan, Sašo Karakatič, Tadej Battelino, Maja Drobnič Radobuljac, 2023, izvirni znanstveni članek

Opis: Background: Our aim was to determine whether child attachment to parents, parent attachment style, and morning cortisol levels were related to diabetes outcomes measured by average glycated hemoglobin (HbA1c), HbA1c variability over 4 years and time in range (TIR) in children with type 1 diabetes (T1D). Research design and methods: 101 children with T1D and one of their parents were assessed at baseline for child attachment (Child Attachment Interview; CAI) and parent attachment (Relationship Structures Questionnaire; ECR-RS). Serum samples were collected for cortisol measurements before the interviews. HbA1c levels were measured during a 4-year follow-up period at regular 3-monthly visits, and data for TIR were exported from blood glucose measuring devices. Multivariate linear regression models were constructed to identify independent predictors of glycemic outcomes. Results: More girls than boys exhibited secure attachment to their mothers. The results of the regression models showed that securely attached girls (CAI) had higher average HbA1c than did insecurely attached girls (B = -0.64, p = 0.03). In boys, the more insecure the parent's attachment style, the worse the child's glycemic outcome: the higher the average Hb1Ac (B = 0.51, p = 0.005), the higher the HbA1c variability (B = 0.017, p = 0.011), and the lower the TIR (B = -8.543, p = 0.002). Conclusions: Attachment in close relationships is associated with glycemic outcomes in children with T1D, and we observed significant differences between sexes. A sex- and attachment-specific approach is recommended when treating children with less favorable glycemic outcomes. Special attention and tailored support should be offered to securely attached girls in transferring responsibility for diabetes care and at least to male children of insecurely attached parents to prevent suboptimal glycemic control. Further studies in larger samples and more daily cortisol measurements may help us better understand the links between stress response, attachment and T1D.
Ključne besede: attachment, childhood and adolescence, cortisol, diabetes control, time in range
Objavljeno v DKUM: 21.05.2024; Ogledov: 98; Prenosov: 5
.pdf Celotno besedilo (980,31 KB)
Gradivo ima več datotek! Več...

4.
DynFS: dynamic genotype cutting feature selection algorithm
Dušan Fister, Iztok Fister, Sašo Karakatič, 2023, izvirni znanstveni članek

Ključne besede: feature selection, nature-inspired algorithms, swarm intelligence, optimization
Objavljeno v DKUM: 05.04.2024; Ogledov: 159; Prenosov: 10
.pdf Celotno besedilo (1,14 MB)

5.
6.
Oblikovanje uporabniške izkušnje z inteligentnim sistemom ChatGPT : magistrsko delo
Elena Osrajnik, 2023, magistrsko delo

Opis: V magistrskem delu je bila raziskana možnost oblikovanja uporabniške izkušnje spletne strani s pomočjo sistema ChatGPT. Zastavljene so bile zahteve, preko katerih je bil voden do odkrivanja načinov ugotavljanja potreb uporabnikov, kreiranja person, testiranja uporabnosti strani, raziskovanja potekov uporabe spletne strani, načinov merjenja in povečanja konverzije strani ter oblikovanja pristajalne strani z uporabo elementov uporabniške izkušnje. Po analizi odgovorov je bilo ugotovljeno, da je ChatGPT sposoben sooblikovati uporabniško izkušnjo fiktivne spletne strani, ki je primerljiva z oblikovanjem, ki bi ga izvedel človek. Spletna stran, za katero je bila ustvarjena uporabniška izkušnja, je bila s sistemom tudi vizualno predstavljena.
Ključne besede: Uporabniška izkušnja, oblikovanje, ChatGPT, umetna inteligenca.
Objavljeno v DKUM: 12.10.2023; Ogledov: 544; Prenosov: 116
.pdf Celotno besedilo (1,54 MB)

7.
Podatkovno podprta evalvacija znanj in spretnosti : magistrsko delo
Damijan Robnik, 2023, magistrsko delo

Opis: Magistrsko delo opisuje uporabo algoritma Node2Vec za analizo odnosov med strokovnjaki in njihovimi izkušnjami na področju informacijske tehnologije (IT). V delu je predstavljen algoritem za generiranje simuliranih izkušenj strokovnjakov, ki se uporabi za ustvarjanje grafa kot vhod v Node2Vec. Prav tako so predstavljeni rezultati ankete, s katero smo pridobili potrebne podatke o izkušnjah strokovnjakov na področju IT. Na podlagi teh podatkov in simuliranih izkušenj je ocenjena uspešnost algoritma Node2Vec pri razvrščanju spletnih programerjev v skupine (gruče).
Ključne besede: IT znanja, teorija grafov, nevronske mreže, Node2Vec
Objavljeno v DKUM: 12.10.2023; Ogledov: 408; Prenosov: 30
.pdf Celotno besedilo (7,14 MB)

8.
Analyzing EEG signal with Machine Learning in Python : graduation thesis
Evgenija Siljanovska, 2023, diplomsko delo

Opis: This thesis presents a comprehensive analysis of EEG data using Python libraries, MNE and machine learning techniques. The thesis focuses on utilizing these tools to extract valuable insights from EEG recordings. Our dataset consists of EEG data in the BrainVision format, acquired during a psychology experiment. The analysis involves preprocessing, filtering, segmentation, and visualization of the EEG data. Additionally, machine learning algorithms are employed to classify and predict patterns within the EEG signals. The findings showcase the effectiveness of Python, MNE, and machine learning in EEG analysis.
Ključne besede: EEG data, MNE, Machine learning, Analyzing
Objavljeno v DKUM: 17.08.2023; Ogledov: 536; Prenosov: 48
.pdf Celotno besedilo (2,66 MB)

9.
Improved Boosted Classification to Mitigate the Ethnicity and Age Group Unfairness
Ivona Colakovic, Sašo Karakatič, 2022, objavljeni znanstveni prispevek na konferenci

Opis: This paper deals with the group fairness issue that arises when classifying data, which contains socially induced biases for age and ethnicity. To tackle the unfair focus on certain age and ethnicity groups, we propose an adaptive boosting method that balances the fair treatment of all groups. The proposed approach builds upon the AdaBoost method but supplements it with the factor of fairness between the sensitive groups. The results show that the proposed method focuses more on the age and ethnicity groups, given less focus with traditional classification techniques. Thus the resulting classification model is more balanced, treating all of the sensitive groups more equally without sacrificing the overall quality of the classification.
Ključne besede: fairness, classification, boosting, machine learning
Objavljeno v DKUM: 02.08.2023; Ogledov: 456; Prenosov: 37
.pdf Celotno besedilo (884,95 KB)
Gradivo ima več datotek! Več...

10.
Proučevanje zunanjih dejavnikov pri napovedovanju cene kriptovalut s strojnim učenjem : diplomsko delo
Jakob Cvetko, 2023, diplomsko delo

Opis: Zmožnost napovedovanja gibanja cene finančnih instrumentov predstavlja priložnost za visoke zaslužke. Eni izmed tehnični pristopov, ki se na področju finančnega trgovanja že dalj časa uspešno uporabljajo, so metode strojnega učenja. V diplomski nalogi smo se ukvarjali z napovedovanjem cene kriptovalute Bitcoin. Modeliranje smo začeli s pridobivanjem raznih podatkov, povezanih s ceno kriptovalute, in nato z algoritmom XGBoost izdelali napovedni model. Razumevanje napovedi je ključnega pomena, zato smo uporabili razlagalni algoritem SHAP, s katerim smo dobili globlji vpogled v napovedni model. Izkazalo se je, da imajo podatki, neposredno vezani na ceno kriptovalute, največjo vlogo pri napovedi, temu pa sledi indeks strahu in pohlepa.
Ključne besede: kriptovalute, strojno učenje, XGBoost, napovedovanje časovnih vrst, SHAP
Objavljeno v DKUM: 07.06.2023; Ogledov: 603; Prenosov: 70
.pdf Celotno besedilo (2,08 MB)

Iskanje izvedeno v 9.31 sek.
Na vrh
Logotipi partnerjev Univerza v Mariboru Univerza v Ljubljani Univerza na Primorskem Univerza v Novi Gorici