Studia Universitatis Babeș-Bolyai Informatica

Matching Apictorial Puzzle Pieces Using Deep Learning

2024-06-10T08:48:32+00:00

Finding matches between puzzle pieces is a difficult problem relevant to applications that involve restoring broken objects. The main difficulty comes from the similarity of the puzzle pieces and the very small difference between a pair of pieces that almost match and one that does. The proposed solution is based on deep learning models and has two steps: firstly, the pieces are segmented from images with a U-Net model; then, matches are found with a Siamese Neural Network. To reach our goal, we created our own dataset, containing 462 images and just as many masks. With these masks, we built 3318 pairs of images, half of them representing pieces that fit together and half that do not. Our most relevant result is estimating correctly for 290 out of 332 pairs whether they match.

Received by the editors: 23 October 2023.

2010 Mathematics Subject Classification. 68T45, 68U10.

1998 CR Categories and Descriptors. I.2.10 Artificial Intelligence: Vision and Scene Understanding – Shape; I.4.6 Image Processing and Computer Vision: Segmentation – Edge and feature detection.

3D Deformable Object Matching Using Graph Neural Networks

2024-06-10T08:57:33+00:00

Considering the current advancements in computer vision it can be observed that most of it is focused on two-dimensional imagery. This includes problems such as classification, regression, and the lesser-known object matching problem. While object matching ca be viewed as a solved problem in a two-dimensional space, for a three-dimensional space there is a long way to go, especially for non-rigid objects. The problem is focused on matching a given object to a target object. We propose a solution based on Graph Neural Networks that tries to generalize over multiple objects at once, based on self-attention and cross-attention blocks for the network. To test our solution, we utilised five convolutional operators for the layers of the model. The convolutional operators we compared included GCNConv, ChebConv, SAGEConv, TAGConv, and FeaStConv. This paper aims to find the best operators for our architecture and the task. Our approach obtained favourable results for predicting the barycentric weights for the model, while struggling to predict the triangle indexes. The best results were obtained for the models using GCNConv, for the triangles index prediction and FeaStConv for the barycentric coordinates prediction.

Received by the editors: 29 November 2023.

2010 Mathematics Subject Classification. 68T45, 68U05.

1998 CR Categories and Descriptors. I.2.10 Artificial Intelligence: Vision and Scene Understanding – 3D/stereo scene analysis; I.3.5 Computer Graphics: Computational Geometry and Object Modeling – Geometric algorithms, languages, and systems.

A Comprehensive Evaluation of Rough Sets Clustering in Uncertainty Driven Contexts

2024-06-10T09:05:47+00:00

This paper presents a comprehensive evaluation of the Agent BAsed Rough sets Clustering (ABARC) algorithm, an approach using rough sets theory for clustering in environments characterized by uncertainty. Several experiments utilizing standard datasets are performed in order to compare ABARC against a range of supervised and unsupervised learning algorithms. This comparison considers various internal and external performance measures to evaluate the quality of clustering. The results highlight the ABARC algorithm’s capability to effectively manage vague data and outliers, showcasing its advantage in handling uncertainty in data. Furthermore, they also emphasize the importance of choosing appropriate performance metrics, especially when evaluating clustering algorithms in scenarios with unclear or inconsistent data.

Received by the editors: 5 March 2024.

2010 Mathematics Subject Classification. 68T37, 68T99.

1998 CR Categories and Descriptors. I.5.3 Pattern Recognition.: Clustering – Algorithms; I.5.3 Pattern Recognition.: Clustering – Similarity measures.

Competitive Influence Maximization in Trust-Based Social Networks With Deep Q-Learning

2024-06-10T09:11:25+00:00

Social network analysis is a rapidly evolving research area having several real-life application areas, e.g. digital marketing, epidemiology, spread of misinformation. Influence maximization aims to select a subset of nodes in such manner that the information propagated over the network is maximized. Competitive influence maximization, which describes the phenomena of multiple actors competing for resources within the same infrastructure, can be solved with a greedy approach selecting the seed nodes utilizing the influence strength between nodes. Recently, deep reinforcement learning methods were applied for estimating the influence strength. We train a controller with reinforcement learning for selecting a node list of given length as the initial seed set for the information spread. Our experiments show that deep Q-learning methods are suitable to analyze the competitive influence maximization on trust and distrust based social networks.

Received by the editors: 1 March 2023.

2010 Mathematics Subject Classification. 68T05.

1998 CR Categories and Descriptors. G.2.2 Discrete Mathematics: Graph Theory – Network problems; G.3. Probability and Statistics: Markov Processes; I.2.6 Artificial Intelligence: Learning – Connectionism and neural nets.

Automatic Detection of Verbal Deception in Romanian With Artificial Intelligence Methods

2024-06-10T09:19:57+00:00

Automatic deception detection is an important task with several applications in both direct physical human communication, as well as in computer-mediated one. The objective of this paper is to study the nature of deceptive language. The primary goal of this study is to investigate deception in Romanian written communication. We created a number of artificial intelligence models (based on Support Vector Machine, Random Forest, and Artificial Neural Network) to detect dishonesty in a topic-specific corpus. To assess the efficiency of the Linguistic Inquiry and Word Count (LIWC) categories in Romanian, we conducted a comparison between multiple text representations based on LIWC, TF-IDF, and LSA. The results show that in the case of datasets with a common subject such as the one we used regarding friendship, text categorization is more successful using general text representations such as TF-IDF or LSA. The proposed approach achieves an accuracy of the classification of 91.3%, outperforming the similar approaches presented in the literature. These findings have implications in fields like linguistics and opinion mining, where research on this subject in languages other than English is necessary.

Received by the editors: 29 April 2024.

2010 Mathematics Subject Classification. 68T50.

1998 CR Categories and Descriptors. I.2.7 ARTIFICIAL INTELLIGENCE.: Nat- ural Language Processing –Text analysis.