<abstract xmlns="http://www.w3.org/1999/xhtml">

We reiterate the theoretical basics of holographic associative memory, and conduct two experiments. During the first experiment, we teach the system many associations, while during the second experiment, we teach it only one association. In both cases, the recalling capability of the system is examined from different aspects.
</abstract>

Experiments with holographic associative memory

<abstract xmlns="http://www.w3.org/1999/xhtml">

Let G be a simple connected graph of order n and size m. The matrix L(G)= D(G)− A(G) is called the Laplacian matrix of the graph G,where D(G) and A(G) are the degree diagonal matrix and the adjacency matrix, respectively. Let the vertex degree sequence be d1 ≥ d2 ≥··· ≥ dn and let μ1 ≥ μ2 ≥··· ≥ μn−1 >μn = 0 be the eigenvalues of the Laplacian matrix of G. The graph invariants, Laplacian energy (LE), the Laplacian-energy-like invariant (LEL) and the Kirchhoff index (Kf), are defined in terms of the Laplacian eigenvalues of graph G, as <inline-formula>
<alternatives>
<inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="graphic/j_ausi-2022-0011_eq_001.png"></inline-graphic>
<math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>L</mi><mi>E</mi><mo>=</mo><msubsup><mo>∑</mo><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mi>n</mi></msubsup><mrow><mrow><mo>|</mo> <mrow><msub><mrow><mi>μ</mi></mrow><mi>i</mi></msub><mo>-</mo><mfrac><mrow><mn>2</mn><mi>m</mi></mrow><mi>n</mi></mfrac></mrow> <mo>|</mo></mrow></mrow></mrow></math>
<tex-math>LE = \sum\nolimits_{i = 1}^n {\left| {{\mu _i} - {{2m} \over n}} \right|}</tex-math>
</alternatives>
</inline-formula>, <inline-formula>
<alternatives>
<inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="graphic/j_ausi-2022-0011_eq_002.png"></inline-graphic>
<math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>L</mi><mi>E</mi><mi>L</mi><mo>=</mo><msubsup><mo>∑</mo><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>n</mi><mo>-</mo><mn>1</mn></mrow></msubsup><mrow><msqrt><mrow><msub><mrow><mi>μ</mi></mrow><mi>i</mi></msub></mrow></msqrt></mrow></mrow></math>
<tex-math>LEL = \sum\nolimits_{i = 1}^{n - 1} {\sqrt {{\mu _i}} }</tex-math>
</alternatives>
</inline-formula> and <inline-formula>
<alternatives>
<inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="graphic/j_ausi-2022-0011_eq_003.png"></inline-graphic>
<math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mrow><mi>K</mi><mi>f</mi><mo>=</mo><mi>n</mi><msubsup><mo>∑</mo><mrow><mi>i</mi><mo>=</mo><mn>1</mn></mrow><mrow><mi>n</mi><mo>-</mo><mn>1</mn></mrow></msubsup><mrow><mfrac><mn>1</mn><mrow><msub><mrow><mi>μ</mi></mrow><mi>i</mi></msub></mrow></mfrac></mrow></mrow></math>
<tex-math>Kf = n\sum\nolimits_{i = 1}^{n - 1} {{1 \over {{\mu _i}}}}</tex-math>
</alternatives>
</inline-formula> respectively. In this paper, we obtain a new bound for the Laplacian-energy-like invariant LEL and establish the relations between Laplacian-energy-like invariant LEL and the Kirchhoff index Kf.Further,weobtain the relations between the Laplacian energy LE and Kirchhoff index Kf.
</abstract>

Computing Laplacian energy, Laplacian-energy-like invariant and Kirchhoff index of graphs

<abstract xmlns="http://www.w3.org/1999/xhtml">

Residual closeness is recently proposed as a vulnerability measure to characterize the stability of complex networks. Residual closeness is essential in the analysis of complex networks, but costly to compute. Currently, the fastest known algorithms run in polynomial time. Motivated by the fast-growing need to compute vulnerability measures on complex networks, new algorithms for computing node and edge residual closeness are introduced in this paper. Those proposed algorithms reduce the running times to Θ(n3) and Θ (n4) on unweighted networks, respectively, where n is the number of nodes.
</abstract>

Computational complexity of network vulnerability analysis

<abstract xmlns="http://www.w3.org/1999/xhtml">

Despite the practical importance of accurate long-term electricity price forecast with high resolution - and the significant need for that - only small percentage of the tremendous papers on energy price forecast attempted to target this topic. Its reason can be the high volatility of electricity prices and the hidden – and often unpredictable – relations with its influencing factors.
In our research, we performed different experiments to predicate hourly Hungarian electricity prices using deep neural networks, for short-term and long-term, too. During this work, investigations were made to compare the results of different network structures and to determine the effect of some environmental factors (meteorologic data and date/time - beside the historical electricity prices). Our results were promising, mostly for short-term forecasts - especially by using a deep neural network with one ConvLSTM encoder.
</abstract>

Hourly electricity price forecast for short-and long-term, using deep neural networks

<abstract xmlns="http://www.w3.org/1999/xhtml">

The automated segmentation of brain tissues and lesions represents a widely investigated research topic. The Brain Tumor Segmentation Challenges (BraTS) organized yearly since 2012 provided standard training and testing data and a unified evaluation framework to the research community, which provoked an intensification in this research field. This paper proposes a solution to the brain tumor segmentation problem, which is built upon the U-net architecture that is very popular in medical imaging. The proposed procedure involves two identical, cascaded U-net networks with 3D convolution. The first stage produces an initial segmentation of a brain volume, while the second stage applies a post-processing based on the labels provided by the first stage. In the first U-net based classification, each pixel is characterized by the four observed features (T1, T2, T1c, and FLAIR), while the second identical U-net works with four features extracted from the volumetric neighborhood of the pixels, representing the ratio of pixels with positive initial labeling within the neighborhood. Statistical accuracy indexes are employed to evaluate the initial and final segmentation of each MRI record. Tests based on BraTS 2019 training data set led to average Dice scores over 87%. The postprocessing step can increase the average Dice scores by 0.5%, it improves more those volumes whose initial segmentation was less successful.
</abstract>

Sapientia Hungarian University of Transylvania

University Research, Innovation and Service Center

A two-stage U-net approach to brain tumor segmentation from multi-spectral MRI records

<abstract xmlns="http://www.w3.org/1999/xhtml">

Automatic bokeh is one of the smartphone’s essential photography effects. This effect enhances the quality of the image where the subject background gets out of focus by providing a soft (i.e., diverse) background. Most smartphones have a single rear camera that is lacking to provide which effects need to be applied to which kind of images. To do the same, smartphones depend on different software to generate the bokeh effect on images. Blur, Color-point, Zoom, Spin, Big Bokeh, Color Picker, Low-key, High-Key, and Silhouette are the popular bokeh effects. With this wide range of bokeh types available, it is difficult for the user to choose a suitable effect for their images. Deep Learning (DL) models (i.e., MobileNetV2, InceptionV3, and VGG16) are used in this work to recommend high-quality bokeh effects for images. Four thousand five hundred images are collected from online resources such as Google images, Unsplash, and Kaggle to examine the model performance. 85% accuracy has been achieved for recommending different bokeh effects using the proposed model MobileNetV2, which exceeds many of the existing models.
</abstract>

Rendering automatic bokeh recommendation engine for photos using deep learning algorithm

<abstract xmlns="http://www.w3.org/1999/xhtml">

In this paper we investigate the comments from the AlgoRythmics YouTube channel using the Comment Term Frequency Comparison social media analytics method. Comment Term Frequency Comparison can be a useful tool to understand how a social media platform, such as a Youtube channel is being discussed by users and to identify opportunities to engage with the audience. Understanding viewer opinions and reactions to a video, identifying trends and patterns in the way people are discussing a particular topic, and measuring the effectiveness of a video in achieving its intended goals is one of the most important points of view for a channel to develop. Youtube comment analytics can be a valuable tool looking to understand how the AlgoRythmics channel videos are being received by viewers and to identify opportunities for improvement. Our study focuses on the importance of user feedback based on ten algorithm visualization videos from the AlgoRythmics channel. In order to find evidence how our channel works and new ideas to improve we used the so-called comment term frequency comparison social media analytics method to investigate the main characteristics of user feedback. We analyzed the comments using both Youtube Studio Analytics and Mozdeh Big Data Analysis tool.
</abstract>

Sapientia Hungarian University of Transylvania Târgu Mureş

Investigating the AlgoRythmics YouTube channel: the Comment Term Frequency Comparison social media analytics method

<abstract xmlns="http://www.w3.org/1999/xhtml">

The Nirmala matrix of a graph and its energy have recently defined. In this paper, we establish some features of the Nirmala eigenvalues. Then we propose various bounds on the Nirmala spectral radius and energy. Moreover, we derive a bound on the Nirmala energy including graph energy and maximum vertex degree.
</abstract>

Bounds on Nirmala energy of graphs

<abstract xmlns="http://www.w3.org/1999/xhtml">

The automatic segmentation of medical images stands at the basis of modern medical diagnosis, therapy planning and follow-up studies after interventions. The accuracy of the segmentation is a key element in assisting the work of the physician, but the efficiency of the process is also relevant. This paper introduces a feature selection strategy that attempts to define reduced feature sets for ensemble learning methods employed in brain tumor segmentation based on MRI data such a way that the segmentation outcome hardly suffers any damage. Initially, the full set of observed and generated features are deployed in ensemble training and prediction on testing data, which provide us information on all couples of features from the full feature set. The extracted pairwise data is fed to a Markov clustering (MCL) algorithm, which uses a graph structure to characterize the relation between features. MCL produces connected subgraphs that are totally separated from each other. The largest such subgraph defines the group of features which are selected for evaluation. The proposed technique is evaluated using the high-grade and low-grade tumor records of the training dataset of the BraTS 2019 challenge, in an ensemble learning framework relying on binary decision trees. The proposed method can reduce the set of features to 30%ofits initial size without losing anything in terms of segmentation accuracy, significantly contributing to the efficiency of the segmentation process. A detailed comparison of the full set of 104 features and the reduced set of 41 features is provided, with special attention to highly discriminative and redundant features within the MRI data.
</abstract>

Computational Intelligence Research Group, Sapientia Hungarian University of Transylvania

A feature selection strategy using Markov clustering, for the optimization of brain tumor segmentation from MRI data

<abstract xmlns="http://www.w3.org/1999/xhtml">

A signed graph is a graph in which positive or negative signs are assigned to its edges. We consider equitable colouring and Hamiltonian colouring to obtain induced signed graphs. An equitable colour-induced signed graph is a signed graph constructed from a given graph in which each edge uv receives a sign (−1)|c(v)−c(u)|,where c is an equitable colouring of vertex v. A Hamiltonian colour-induced signed graph is a signed graph obtained from a graph G in which for each edge e = uv, the signature function σ(uv)=(−1)|c(v)−c(u)|, gives a sign such that, |c(u)− c(v)| ≥ n − 1 − D(u, v) where c is a function that assigns a colour to each vertex satisfying the given condition. This paper discusses the properties and characteristics of signed graphs induced by the equitable and Hamiltonian colouring of graphs.
</abstract>

niranjana.sudheer@maths.christuniversity.in

Some new results on colour-induced signed graphs

AHEAD OF PRINT

Volume 15 (2023): Issue 2 (December 2023)

Volume 15 (2023): Issue 1 (August 2023)

Volume 14 (2022): Issue 2 (December 2022)

Volume 14 (2022): Issue 1 (August 2022)

Volume 13 (2021): Issue 2 (December 2021)

Volume 13 (2021): Issue 1 (June 2021)

Volume 12 (2020): Issue 2 (December 2020)

Volume 12 (2020): Issue 1 (July 2020)

Volume 11 (2019): Issue 2 (December 2019)

Volume 11 (2019): Issue 1 (August 2019)

Volume 10 (2018): Issue 2 (December 2018)

Volume 10 (2018): Issue 1 (August 2018)

Volume 9 (2017): Issue 2 (December 2017)

Volume 9 (2017): Issue 1 (July 2017)

Volume 8 (2016): Issue 2 (December 2016)

Volume 8 (2016): Issue 1 (June 2016)

Volume 7 (2015): Issue 2 (December 2015)

Volume 7 (2015): Issue 1 (June 2015)

Volume 6 (2014): Issue 2 (December 2014)

Volume 6 (2014): Issue 1 (June 2014)

Volume 5 (2013): Issue 2 (December 2013)

Volume 5 (2013): Issue 1 (July 2013)

Acta Universitatis Sapientiae, Informatica

Acta Universitatis Sapientiae Informatica publishes peer-reviewed, high-quality original papers and surveys in English, in all areas of Computer Science and allied fields. The papers must be self-contained, including the purpose of the research, the basic definitions, the antecedents of the research, the presentation of the state-of-the-art and novel contributions. The articles must contain sufficient references to give the reader a broad view of the studied subject.  Articles must follow the technical requirements described in the Information for Authors. Please note, that if an article fails to do so, it will be rejected without peer-review.  The corresponding author is responsible for obtaining the permission of coauthors and of the authorities or institutions, if needed, for publication, the Editorial Board is disclaiming any responsibility.  The papers have to be prepared carefully, and must include the relevant keywords. References should be listed alphabetically based on the Instruction for Authors given at the address www.acta.sapientia.ro/acta-info/informatica-main.htm.  Why subscribe and read  The journal publishes papers in all fields of Computer Science, so it can satisfy a lot of interests; the online version is very attractive with the hyperlinks.  Why submit  The issues appear regularly 2 times per year, the waiting time is not too long, the online version contains hyperlinks.  Rejection Rate   56%   Archiving  Sciendo archives the contents of this journal in Portico - digital long-term preservation service of scholarly books, journals and collections.  Plagiarism Policy  The editorial board is participating in a growing community of Similarity Check System's users in order to ensure that the content published is original and trustworthy. Similarity Check is a medium that allows for comprehensive manuscripts screening, aimed to eliminate plagiarism and provide a high standard and quality peer-review process.  Article Processing Charge (APC)  The journal does not have article processing charges (APCs) nor article submission charges. 