Mining High Utility Itemsets with Hill Climbing and Simulated Annealing

High utility itemset mining (HUIM) is the task of finding all items set, purchased together, that generate a high profit in a transaction database. In the past, several algorithms have been developed to mine high utility itemsets (HUIs). However, most of them cannot properly handle the exponential search space while finding HUIs when the size of the database and total number of items increases. Recently, evolutionary and heuristic algorithms were designed to mine HUIs, which provided considerable performance improvement. However, they can still have a long runtime and some may miss many HUIs. To address this problem, this article proposes two algorithms for HUIM based on Hill Climbing (HUIM-HC) and Simulated Annealing (HUIM-SA). Both algorithms transform the input database into a bitmap for efficient utility computation and for search space pruning. To improve population diversity, HUIs discovered by evolution are used as target values for the next population instead of keeping the current optimal values in the next population. Through experiments on real-life datasets, it was found that the proposed algorithms are faster than state-of-the-art heuristic and evolutionary HUIM algorithms, that HUIM-SA discovers similar HUIs, and that HUIM-SA evolves linearly with the number of iterations.

Electricity Substitution Potential Prediction Based on Tent-CSO-CG-SSA-Improved SVM—A Case Study of China

Nowadays, fossil energy continues to dominate China’s energy usage; its inefficient use and large crude emissions of coal and fuel oil in its end-consumption have brought about great pressure to reduce emissions. Electrical power substitution as a development strategy is an important step toward achieving sustainable development, the transformation of the end-use energy consumption structure, and double carbon goals. To better guide the broad promotion of electrical power substitution, and to offer theoretical support for its development, this paper quantifies the amount of electrical power substitution and the influencing factors that affect the potential of electrical energy substitution. This paper proposes a hybrid model, combining Tent chaos mapping (Tent), chicken swarm optimization (CSO), Cauchy–Gaussian mutation (CG), the sparrow search algorithm (SSA), and a support vector machine (SVM), as a Tent-CSO-CG-SSA-SVM model, which first uses the method of Tent chaos mapping to initialize the sparrow population in order to increase population diversity and improve the search ability of the algorithm. Then, the CSO is introduced to update the positions of sparrows, and the CG method is introduced to make the algorithm jump out of the local optimum, in order to improve the global search ability of the SSA. Finally, the final electrical power substitution potential prediction model is obtained by optimizing the SVM through a multi-algorithm combination approach. To verify the validity of the model, two regions in China were used as case studies for the prediction analysis of electrical energy substitution potential, and the prediction results were compared with multiple models. The results of the study show that Tent-CSO-CG-SSA-SVM offers a good improvement in prediction accuracy, and that Tent-CSO-CG-SSA-SVM is a promising method for the prediction of electrical power substitution potential.

The diversity of Passalora fulva Isolates Collected from Tomato Plants in US high tunnels

High tunnels extend the growing season of high value crops, including tomatoes, but the environmental conditions within high tunnels favor the spread of the tomato leaf mold pathogen, Passalora fulva (syn. Cladosporium fulvum). Tomato leaf mold results in defoliation, and if severe, losses in yield. Despite substantial research, little is known regarding the genetic structure and diversity of populations of P. fulva associated with high tunnel tomato production in the United States. From 2016 to 2019, a total of 50 P. fulva isolates were collected from tomato leaf samples in high tunnels in the Northeast and Minnesota. Other Cladosporium species were also isolated from the leaf surfaces. Koch’s postulates were conducted to confirm that P. fulva was the cause of the disease symptoms observed. Race determination experiments revealed that the isolates belonged to either race 0 (six isolates) or race 2 (44 isolates). Polymorphisms were identified within four previously characterized effector genes Avr2, Avr4, Avr4e, and Avr9. The largest number of polymorphisms were observed for Avr2. Both mating type genes, MAT1-1-1 and MAT1-2-1, were present in the isolate collection. For further insights into the pathogen diversity, the 50 isolates were genotyped at 7,514 single-nucleotide polymorphism loci using genotyping-by-sequencing: differentiation by region but not by year was observed. Within the collection of 50 isolates, there were 18 distinct genotypes. Information regarding P. fulva population diversity will enable better management recommendations for growers, as high tunnel production of tomatoes expands.

Educator Perceptions of Early Learning Environments as Places for Privileging Social Justice in Rural and Remote Communities

Early childhood education and care (ECEC) in Australia has long been associated with the concept of social justice, however, a clear understanding of what it looks like across diverse services and communities is not available. This article reports the process of inquiry, as well as the outcomes, of a small-scale study designed to uncover the perceptions of ECEC educators working in rural and remote communities in the state of Queensland. Data were collected through individual semi-structured interviews with five educators from rural and remote settings identified as areas experiencing significant growth in population diversity. An initial thematic analysis of the data revealed three key themes. A secondary analysis using a place and space conceptual framework uncovered deeper, more sophisticated meanings of the educator experience of social justice. The research is important in bringing pedagogical conversations to the forefront regarding ECEC educator perceptions of their role in creating a socially just learning environment. In addition to identifying future research possibilities, implications from the findings indicate opportunities for re-examining and rethinking initial teacher education and ongoing professional learning.

Deep learning and model personalization in sensor-based human activity recognition

Human activity recognition (HAR) is a line of research whose goal is to design and develop automatic techniques for recognizing activities of daily living (ADLs) using signals from sensors. HAR is an active research filed in response to the ever-increasing need to collect information remotely related to ADLs for diagnostic and therapeutic purposes. Traditionally, HAR used environmental or wearable sensors to acquire signals and relied on traditional machine-learning techniques to classify ADLs. In recent years, HAR is moving towards the use of both wearable devices (such as smartphones or fitness trackers, since they are daily used by people and they include reliable inertial sensors), and deep learning techniques (given the encouraging results obtained in the area of computer vision). One of the major challenges related to HAR is population diversity, which makes difficult traditional machine-learning algorithms to generalize. Recently, researchers successfully attempted to address the problem by proposing techniques based on personalization combined with traditional machine learning. To date, no effort has been directed at investigating the benefits that personalization can bring in deep learning techniques in the HAR domain. The goal of our research is to verify if personalization applied to both traditional and deep learning techniques can lead to better performance than classical approaches (i.e., without personalization). The experiments were conducted on three datasets that are extensively used in the literature and that contain metadata related to the subjects. AdaBoost is the technique chosen for traditional machine learning, while convolutional neural network is the one chosen for deep learning. These techniques have shown to offer good performance. Personalization considers both the physical characteristics of the subjects and the inertial signals generated by the subjects. Results suggest that personalization is most effective when applied to traditional machine-learning techniques rather than to deep learning ones. Moreover, results show that deep learning without personalization performs better than any other methods experimented in the paper in those cases where the number of training samples is high and samples are heterogeneous (i.e., they represent a wider spectrum of the population). This suggests that traditional deep learning can be more effective, provided you have a large and heterogeneous dataset, intrinsically modeling the population diversity in the training process.

Automating microsatellite screening and primer design from multi-individual libraries using Micro-Primers

Analysis of intra- and inter-population diversity has become important for defining the genetic status and distribution patterns of a species and a powerful tool for conservation programs, as high levels of inbreeding could lead into whole population extinction in few generations. Microsatellites (SSR) are commonly used in population studies but discovering highly variable regions across species’ genomes requires demanding computation and laboratorial optimization. In this work, we combine next generation sequencing (NGS) with automatic computing to develop a genomic-oriented tool for characterizing SSRs at the population level. Herein, we describe a new Python pipeline, named Micro-Primers, designed to identify, and design PCR primers for amplification of SSR loci from a multi-individual microsatellite library. By combining commonly used programs for data cleaning and microsatellite mining, this pipeline easily generates, from a fastq file produced by high-throughput sequencing, standard information about the selected microsatellite loci, including the number of alleles in the population subset, and the melting temperature and respective PCR product of each primer set. Additionally, potential polymorphic loci can be identified based on the allele ranges observed in the population, to easily guide the selection of optimal markers for the species. Experimental results show that Micro-Primers significantly reduces processing time in comparison to manual analysis while keeping the same quality of the results. The elapsed times at each step can be longer depending on the number of sequences to analyze and, if not assisted, the selection of polymorphic loci from multiple individuals can represent a major bottleneck in population studies.

Optimization of the transmission characteristics of an HMCVT for a high-powered tractor based on an improved NSGA-II algorithm

The hydro-mechanical continuously variable transmission (HMCVT) is a critical component of the power transmission system in a tractor. However, the complexity of the operating conditions imposes high requirements on the transmission characteristics. To improve the powerful performance and economy of HMCVTs and satisfy the operational demands of high-powered tractors, a new optimization design method for the characteristic parameters of an HMCVT is proposed. First, the characteristics of an HMCVT are modeled, and the influence of the structural parameters on the transmission characteristics is analyzed. Then, HMCVT performance evaluation indexes are formulated. In accordance with the speed regulation of system, power performance, and economy characteristics, a multi-objective optimization mathematical model is established, and an improved fast non-dominated sorting genetic algorithm (INSGA-II) is designed. The introduction of a normal distribution crossover operator (NDX) and an improved adaptive adjustment mutation operator not only ensures the population diversity but also improves the Pareto solution convergence properties during the process of genetic evolution. The superiority of INSGA-II is verified by comparison with a traditional multi-objective genetic algorithm. Finally, the optimization results show that the torque ratio is increased by approximately 2.81%, 14.32%, 2.31%, and 15.07% in HM1, HM2, HM3, and HM4 respectively. The transmission efficiency is increased by approximately 3.48% and 1.97% in HM1 (HM3) and HM2 (HM4). Also, INSGA-II finds the optimal solution with a faster speed and shorter optimization time than MULGA. This research can serve as a reference for the design and optimization of HMCVTs for high-powered tractors.

Identification of Pueraria spp. through DNA barcoding and comparative transcriptomics

Background Kudzu is a term used generically to describe members of the genus Pueraria. Kudzu roots have been used for centuries in traditional Chinese medicine in view of their high levels of beneficial isoflavones including the unique 8-C-glycoside of daidzein, puerarin. In the US, kudzu is seen as a noxious weed causing ecological and economic damage. However, not all kudzu species make puerarin or are equally invasive. Kudzu remains difficult to identify due to its diverse morphology and inconsistent nomenclature. Results We have generated sequences for the internal transcribed spacer 2 (ITS2) and maturase K (matK) regions of Pueraria montana lobata, P. montana montana, and P. phaseoloides, and identified two accessions previously used for differential analysis of puerarin biosynthesis as P. lobata and P. phaseoloides. Additionally, we have generated root transcriptomes for the puerarin-producing P. m. lobata and the non-puerarin producing P. phaseoloides. Within the transcriptomes, microsatellites were identified to aid in species identification as well as population diversity. Conclusions The barcode sequences generated will aid in fast and efficient identification of the three kudzu species. Additionally, the microsatellites identified from the transcriptomes will aid in genetic analysis. The root transcriptomes also provide a molecular toolkit for comparative gene expression analysis towards elucidation of the biosynthesis of kudzu phytochemicals.

An improved arithmetic optimization algorithm with forced switching mechanism for global optimization problems

<abstract> <p>Arithmetic optimization algorithm (AOA) is a newly proposed meta-heuristic method which is inspired by the arithmetic operators in mathematics. However, the AOA has the weaknesses of insufficient exploration capability and is likely to fall into local optima. To improve the searching quality of original AOA, this paper presents an improved AOA (IAOA) integrated with proposed forced switching mechanism (FSM). The enhanced algorithm uses the random math optimizer probability (<italic>RMOP</italic>) to increase the population diversity for better global search. And then the forced switching mechanism is introduced into the AOA to help the search agents jump out of the local optima. When the search agents cannot find better positions within a certain number of iterations, the proposed FSM will make them conduct the exploratory behavior. Thus the cases of being trapped into local optima can be avoided effectively. The proposed IAOA is extensively tested by twenty-three classical benchmark functions and ten CEC2020 test functions and compared with the AOA and other well-known optimization algorithms. The experimental results show that the proposed algorithm is superior to other comparative algorithms on most of the test functions. Furthermore, the test results of two training problems of multi-layer perceptron (MLP) and three classical engineering design problems also indicate that the proposed IAOA is highly effective when dealing with real-world problems.</p> </abstract>