Features selection by genetic algorithm optimization with k-nearest neighbour and learning ensemble to predict Parkinson disease

<span lang="EN-US">Among the several ways followed for detecting Parkinson's disease, there is the one based on the speech signal, which is a symptom of this disease. In this paper focusing on the signal analysis, a data of voice records has been used. In these records, the patients were asked to utter vowels “a”, “o”, and “u”. Discrete wavelet transforms (DWT) applied to the speech signal to fetch the variable resolution that could hide the most important information about the patients. From the approximation a3 obtained by Daubechies wavelet at the scale 2 level 3, 21 features have been extracted: a <a name="_Hlk88480766"></a>linear predictive coding (LPC), energy, zero-crossing rate (ZCR), mel frequency cepstral coefficient (MFCC), and wavelet Shannon entropy. Then for the classification, the K-nearest neighbour (KNN) has been used. The KNN is a type of instance-based learning that can make a decision based on approximated local functions, besides the ensemble learning. However, through the learning process, the choice of the training features can have a significant impact on overall the process. So, here it stands out the role of the genetic algorithm (GA) to select the best training features that give the best accurate classification.</span>

Dominating lengthscales of zebrafish collective behaviour

Collective behaviour in living systems is observed across many scales, from bacteria to insects, to fish shoals. Zebrafish have emerged as a model system amenable to laboratory study. Here we report a three-dimensional study of the collective dynamics of fifty zebrafish. We observed the emergence of collective behaviour changing between ordered to randomised, upon adaptation to new environmental conditions. We quantify the spatial and temporal correlation functions of the fish and identify two length scales, the persistence length and the nearest neighbour distance, that capture the essence of the behavioural changes. The ratio of the two length scales correlates robustly with the polarisation of collective motion that we explain with a reductionist model of self–propelled particles with alignment interactions.

Machine learning of phase transitions in nonlinear polariton lattices

Polaritonic lattices offer a unique testbed for studying nonlinear driven-dissipative physics. They show qualitative changes of their steady state as a function of system parameters, which resemble non-equilibrium phase transitions. Unlike their equilibrium counterparts, these transitions cannot be characterised by conventional statistical physics methods. Here, we study a lattice of square-arranged polariton condensates with nearest-neighbour coupling, and simulate the polarisation (pseudospin) dynamics of the polariton lattice, observing regions with distinct steady-state polarisation patterns. We classify these patterns using machine learning methods and determine the boundaries separating different regions. First, we use unsupervised data mining techniques to sketch the boundaries of phase transitions. We then apply learning by confusion, a neural network-based method for learning labels in a dataset, and extract the polaritonic phase diagram. Our work takes a step towards AI-enabled studies of polaritonic systems.

Comparison of CNN Algorithms for Feature Extraction on Fundus Images to Detect Glaucoma

Glaucoma is a disease where the optic nerve of the eyes is smashed up due to the building up of pressure inside the vision point. This has no symptoms at the initial stages, and hence, patients with this disease cannot identify them at the beginning stage. It is explained as if the pressure in the eye increases, then it will hurt the optic nerve which sends images to the brain. This will lead to permanent vision loss or total blindness. The existing method used for the detection of glaucoma includes k-nearest neighbour and support vector machine algorithms. The k-nearest neighbour algorithm and support vector machine algorithm are the machine learning methods for both categorization and degeneration problems. The drawback in using these algorithms is that we can get accuracy level only up to 80%. The proposed methods in this study focus on the convolution neural network for the recognition of glaucoma. In this study, 2 architectures of VGG, Inception method, AlexNet, GoogLeNet, and ResNet architectures which provide accuracy levels up to 100% are presented.

Pengembangan Aplikasi Penentuan Rute Pengiriman Barang Berdasarkan Berat dan Time Windows Menggunakan Metode Nearest Neighbour dan Tabu Search

Salah satu contoh praktis dari CVRPTW adalah pengiriman barang. Faktor penting dalam pengiriman barang adalah biaya, kecepatan, pelayanan dan konsistensi. Agar faktor-faktor tersebut terpenuhi secara optimal harus diperhatikan muatan barang serta time windows. Muatan berpengaruh pada faktor pelayanan dan konsistensi, sehingga harus dipilih rute yang tepat dengan jarak terpendek serta ketepatan kapasitas barang. Time windows berpengaruh pada faktor kecepatan dan biaya pengiriman sehingga pengiriman barang harus dilakukan sesuai waktu yang ditentukan dan jam operasional perusahaan. Penelitian ini bertujuan menghasilkan rute pengiriman barang yang memperhatikan kapasitas muatan dan waktu tempuh pengiriman. Terdapat dua tahapan penyelesaian yaitu tahap clustering dan pencarian rute optimal. Tahap clustering menggunakan sudut polar dan tpencarian rute optimal menggunakan metode nearest neighbour serta tabu search. Hasil pengujian menunjukkan bahwa rute pengiriman yang dihasilkan oleh sistem dapat melakukan efisiensi jarak pengiriman sebesar 12,18%, waktu pengiriman sebesar 5,54%, muatan sebesar 1,27% dan efisiensi biaya sebesar 12,18%.

Lemonia batavorum sp. nov. from the Netherlands, an overlooked sibling of L. dumi (Lepidoptera: Brahmaeidae)

Barcoding of Dutch specimens of Lemonia dumi (Linnaeus, 1761) (Lepidoptera: Brahmaeidae) and a study of a large collection material of this species resulted in discovery of a new, hitherto undescribed species Lemonia batavorum sp. nov. Based on comparison with L. dumi, species level of L. batavorum sp. nov. is supported by the differences in its external appearance, diff erences in genitalia of both sexes and by 1.92% p-distance to L. dumi as the nearest neighbour. Photographs of specimens and genitalia of both sexes are given.

Enhancement of superexchange due to synergetic breathing and hopping in corner-sharing cuprates

Cuprates with corner-sharing CuO4 plaquettes have received much attention owing to the discoveries of high-temperature superconductivity and exotic states where spin and charge or spin and orbital degrees of freedom are separated. In these systems spins are strongly coupled antiferromagnetically via superexchange mechanisms, with high nearest-neighbour coupling varying among different compounds. The electronic properties of cuprates are also known to be highly sensitive to the presence, distance and displacement of apical oxygens perpendicular to the CuO2 planes. Here we present ab initio quantum chemistry calculations of the nearest-neighbour superexchange antiferromagnetic (AF) coupling J of two cuprates, Sr2CuO3 and La2CuO4. The former lacks apical oxygens, whilst the latter contain two apical oxygens per CuO2 unit completing a distorted octahedral environment around each Cu atom. Good agreement is obtained with experimental estimates for both systems. Analysis of the correlated wavefunctions together with extended superexchange models shows that there is an important synergetic effect of the Coulomb interaction and the O–Cu hopping, namely a correlated breathing-enhanced hopping mechanism. This is a new ingredient in superexchange models. Suppression of this mechanism leads to drastic reduction in the AF coupling, indicating that it is of primary importance in generating the strong interactions. We also find that J increases substantially as the distance between Cu and apical O is increased.

Path Planning for Bulldozers with Curvature Constraints

There is presently no solution to the problem of an autonomous bulldozer pushing mounds of material to desired goal locations in the presence of obstacles whilst obeying the kinematic constraints of the bulldozer. Past work has solved some aspects of this problem, but not all. This research presents the first complete, practical solution to the problem. It works by creating a fixed RRT in advance, and then during operation connecting pushing poses into this RRT using Bezier curves. The RRT algorithm leverages a novel data structure for performing nearest neighbour comparisons for Ackermann-steering vehicles; termed the Distmetree. The resulting pushing states are searched using greedy heuristic search to find a solution and the final path is smoothed with cubic Bezier curves. The mode of operation chosen for best performance also constructs bidirectional RRTs to reach difficult to access pushing poses. The final mode of the algorithm was tested in simulation and proven to be able to solve a wide variety of maps in a few minutes while obeying bulldozer kinematic constraints. The algorithm, whilst not optimal, is complete which is the more desirable property in industry, and the solutions it produces are both feasible and reasonable.