Static Set Research Articles

Ungrading

Abstract The word ungrading means raising an eyebrow at grades as a systemic practice, distinct from simply not grading. The word is a present participle, an ongoing process, not a static set of practices. Too many approaches to grades treat students as if they are interchangeable and fail to recognize their complexity. Educational institutions need to start by rewriting policies and imagining new ways forward for the most marginalized students. This essay examines contemporary approaches to assessment; considers the history of grades; interrogates the bias inherent in standardized systems; and explores methods and approaches for designing assessments that push back against traditional notions of grading.

Ultrasound assessment of Achilles enthesitis: a dedicated training program.

To describe an intensive and multimodal ultrasound (US) training program focused on Achilles enthesitis and to illustrate the learning curve of trainees without experience. Three medical students (trainees) and two rheumatologists experienced in musculoskeletal US (trainers) were involved in the training program, which encompassed one preliminary theoretical-practical meeting and five scanning sessions (two patients per session). The students and one expert performed the US examination of the Achilles enthesis bilaterally. The trainees acquired representative images and assessed the presence of Outcome Measures in Rheumatology (OMERACT) US abnormalities of enthesitis. The experts provided feedback addressing trainees' misinterpretations, and the quality of the acquired images was evaluated. A dedicated questionnaire was used to evaluate the students' confidence. After each session, five sets of static images (total=100 images of most commonly scanned entheses) were provided and scored by the students according to OMERACT US definitions. Total agreement and prevalence and bias adjusted kappa (PABAK) were used to evaluate the concordance between the trainees and the expert sonographer. The total agreement and PABAK significantly improved between the first and fifth scanning sessions (76.2% versus 92.9%, p<0.01, and 0.5 versus 0.79, p<0.01) and between the first and fifth static image sets (64.5% versus 81.9%, p<0.01, and 0.29 versus 0.74, p<0.01). Image quality did not significantly improve (p=0.34). A significant increase in trainees' confidence was registered (p<0.01). The described training program rapidly improved the students' performance in the US assessment of Achilles enthesitis, appearing to be an effective starting model for the future development of pathology-oriented teaching programs for the US in rheumatology.

High frequency monitoring of carbon (iv) oxide evolution from a tropical soil amended with organomineral fertilizer in a screenhouse Ile-Ife, Nigeria

The study compared responses of carbon (iv) oxide evolved from an Ultisol sampled at different time scales and the soil properties when amended with organomineral fertilizer (OMF); thus tracking the carbon footprint of the amendment in ambient air. The carbon (iv) oxide efflux was measured titrimetrically in a static chamber set up in a screenhouse at sub-daily, daily and weekly time scales. At the same time, the antecedent soil properties were determined using standard methods. The treatments involved OMF addition at three rates (0 kg OMF, 40 kg Urea-N + 2.5 tons/ha organic fertilizer and 40 kg Urea-N + 5 tons/ha organic fertilizer) based on some Nigerian indigenous vegetable nutrient requirements. The result showed that a more detailed and accurate understanding of CO2 evolution processes was revealed at daily sampling resolution while sub-daily variation occurred in response to sub-daily variation in ambient temperature. Organomineral fertilizer at 40 kg/ha Urea + 2.5 tons/ha OF rate was comparatively safer in C management through the reduction in the CO2 released thus constituting a better alternative in terms of the greenhouse effect. Thus, the study recommends at least daily monitoring for a detailed understanding of CO2 evolution dynamics and management and identifies OMF rate with less CO2 release as a comparatively better alternative in the production of those vegetables in the environmental context.

MP-GUARD: A novel multi-pronged intrusion detection and mitigation framework for scalable SD-IoT networks using cooperative monitoring, ensemble learning, and new P4-extracted feature set

The ever-increasing complexity of the Internet of Things (IoT) environment demands robust and adaptable intrusion detection frameworks, as existing approaches struggle with real-time traffic analysis, limited scalability, and static feature sets. This paper introduces MP-GUARD, a novel framework that leverages Software-Defined Networking (SDN), machine learning (ML), and a multi-controller architecture to address these challenges. MP-GUARD tackles multi-pronged intrusion attacks in IoT networks by offering real-time intrusion detection, collaborative traffic monitoring, and multi-layered attack mitigation. It achieves this through two core modules: P4-Assisted Cooperative Traffic Monitoring (CTM-P4) and Multi-Pronged Intrusion Detection and Mitigation (MPIDM). CTM-P4 facilitates real-time communication among multiple controllers, enabling dynamic feature extraction leveraging the interconnected state tables within P4-enabled switches. This module introduces a new 22-feature set (12 extracted and 10 computed) for comprehensive network analysis. MPIDM leverages the detailed network insights from CTM-P4 for attack identification and prevention. It introduces Stacked Ensemble Learning with Dynamic P4-Based Feature Selection (SELDP4-FS), achieving exceptional performance with 99.32 % accuracy, 99.24 % F1-score, and 0.49 % false positive rate. Additionally, MPIDM boasts efficient response and detection times of 16ms and 11ms, respectively. Beyond accuracy, MP-GUARD demonstrates significant advantages in terms of scalability and efficiency. The multi-controller architecture offers a 65 % reduction in overhead compared to single-controller setups. Furthermore, this work introduces the Mean Accuracy Steadiness Level (MASL) metric to assess model stability under varying traffic conditions. By combining P4-based feature extraction, dynamic feature selection, cooperative monitoring, ensemble learning, and a multi-controller architecture, MP-GUARD presents a significant contribution to IoT security, offering a scalable and adaptable solution for securing future SD-IoT deployments against evolving threats.

Manual Therapy Versus Exercise Therapy in Patients with Osteoarthritis of Knee -A Comparative Study

AIM OF THE STUDY: The aim of the study to compare the effectiveness of manual therapy versus exercise therapy in patients with knee osteoarthritis in terms of decreasing pain HYPOTHESIS • Null Hypothesis: There may be no significant difference in the effectiveness of manual therapy and exercise therapy in patients with knee osteoarthritis • Alternate Hypothesis: There may be significant difference in the effectiveness of manual therapy and exercise therapy in patients with knee osteoarthritis INTERVENTION Group A: MANUAL THERAPY Manual physiotherapy protocol consisted of Treatment is done for two sessions per week for four weeks. Knee Mandatory interventions: 1. Knee flexion, non-thrust Antero-posterior directed force to the tibia, 2. Tibio-femoral joint ,non-thrust 3. Knee extension, non-thrust Postero-anterior directed force to the tibia 4. Tibio-femoral joint, non-thrust 5. Patellar gliding force, non-thrust 6. Manual stretch to quadriceps, hamstring, triceps surae muscle groups 7. Soft tissue manipulation ,quadriceps and Peripatellar connective tissue, hamstring, hip adductor and triceps surae muscles groups Treatment is done for two sessions per week for four weeks. Group B: EXERCISE THERAPY This group performed a closely supervised standardized knee exercise program at twice a week for four weeks at eight treatment sessions. This program consists of active range of motion exercise for knee, muscle strengthening exercise for hip and knee, muscle stretching for lower limbs and riding a stationary bike. Aerobic exercises • Up to 10 minutes stationary bicycle or walk. Stretching exercises: • Standing calf stretch – 3 repetitions with 30 sec hold • Supine hamstring stretch -3 repetitions with 30 sec hold • Prone quadriceps stretch –3 repetitions with 30 sec hold Range of motion exercises: • In long sitting position, • Knee mid flexion to end range extension – two 30 sec bouts with 3 sec hold at end range • In long sitting position, knee mid flexion to end range flexion – two sec bouts with 3 sec hold at end range Strengthening exercises: • Static quad sets in knee extension –one set of 10 repetitions with 6 sec hold,10 sec rest between repetitions. • Short arc terminal extension exercises for the knee joint. • Quadriceps non-weight bearing progressive resistance exercises with weighted cuffs. • Supine Straight Leg Rise (SLR) Neuro muscular control exercises: Closed chain progression: • Seated leg press – one 30 sec bout • Standing weight –shifting exercises • Side- stepping • Step ups • Forward-backward and shuttle – walking drills RESULTS: Knee function of all patients increased significantly after the 4-week intervention program in both the groups of MANUAL THERAPY and EXERCISE THERAPY but Group A Manual therapy improved more significantly when compared to Group B Exercise therapy. CONCLUSION: Treatment effect was more in the patients received Manual therapy when compared to Exercise therapy. This finding suggest that Manual therapy can be useful intervention supplement in patients with osteoarthritis of knee.

Preparation of ZnO Thick Films Activated with UV-LED for Efficient H2S Gas Sensing

In this work, ZnO thick films were synthesized via two simple and easy methods, mechanochemical synthesis and screen-printing deposition. The ZnO powders were obtained through milling at low temperature with milling times of 20, 40, and 60 min. The ZnO thick films were fabricated by depositing 10 cycles of ZnO inks onto glass substrates. The characterization of ZnO thick films revealed a thickness ranging from 4.9 to 5.4 µm with a surface roughness between 85 and 88 nm. The structural analysis confirmed a hexagonal wurtzite crystalline structure of ZnO, both in powders and in thick films, with a preferred orientation on the (002) and (101) planes. Nanostructures with sizes ranging from 36 to 46 nm were observed, exhibiting irregular agglomerated shapes, with an energy band found between 2.77 and 3.02 eV. A static experimental set up was fabricated for gas sensing tests with continuous UV-LED illumination. The ZnO thick films, well adhered to the glass substrate, demonstrated high sensitivity and selectivity to H2S gas under continuous UV-LED illumination at low operating temperatures ranging from 35 to 80 °C. The sensitivity was directly proportional, ranging from 3.93% to 22.40%, when detecting H2S gas concentrations from 25 to 600 ppm.

Controlling the fold: proprioceptive feedback in a soft origami robot.

We demonstrate proprioceptive feedback control of a one degree of freedom soft, pneumatically actuated origami robot and an assembly of two robots into a two degree of freedom system. The base unit of the robot is a 41 mm long, 3-D printed Kresling-inspired structure with six sets of sidewall folds and one degree of freedom. Pneumatic actuation, provided by negative fluidic pressure, causes the robot to contract. Capacitive sensors patterned onto the robot provide position estimation and serve as input to a feedback controller. Using a finite element approach, the electrode shapes are optimized for sensitivity at larger (more obtuse) fold angles to improve control across the actuation range. We demonstrate stable position control through discrete-time proportional-integral-derivative (PID) control on a single unit Kresling robot via a series of static set points to 17 mm, dynamic set point stepping, and sinusoidal signal following, with error under 3 mm up to 10 mm contraction. We also demonstrate a two-unit Kresling robot with two degree of freedom extension and rotation control, which has error of 1.7 mm and 6.1°. This work contributes optimized capacitive electrode design and the demonstration of closed-loop feedback position control without visual tracking as an input. This approach to capacitance sensing and modeling constitutes a major step towards proprioceptive state estimation and feedback control in soft origami robotics.

Constant static pressure strategy for residential kitchen exhaust system: Deducing control parameters from required exhaust rate and non-distributed data transmission

Roof fans have gradually replaced the non-powered hood as supplementary power to increase exhaust rate in high-rise residential kitchen shafts. However, an effective control strategy for roof fans is lacking. This paper studies the constant static pressure strategy and deduces its control parameters based on required exhaust rate and non-distributed data transmission. First, the static pressure set value (SV) is investigated under the worst exhaust condition, considering variations in shaft heights and measuring positions (MP). Then, an on/off mechanism is set up based on the measured static pressure and the fan frequency when the exhaust rate is adequate. Results show that under the recommended shaft sizes and design concurrent usage rates, the SVs for users' exhaust rate reaching 500 m³/h are −60, −60, and 0 Pa for 10/18/26-floor shafts when the MP is at a shaft's outlet, and roof fans are turned on when the static pressure reaches 100, 90, and 90 Pa and turned off when the fan frequency is below 32, 30, and 22 Hz. The MP at the shafts' outlet can keep the shaft at a slightly negative pressure. Control parameters for specific applications can be obtained through the proposed calculation process. An 18-floor shaft study found that the constant static pressure strategy can achieve 50 % energy savings compared to the constant frequency operation of the roof fan. These findings contribute to reliable control without users' distributed data transmission and to lower energy consumption of the residential kitchen exhaust system.

Active Ankle–Foot Orthosis Design and Computer Simulation with Multi-Objective Parameter Optimization

There are many current active orthosis designs to assist with disabilities related to foot-drop, with most of them intervening during the whole gait cycle. We propose that, for the treatment of foot-drop, it is possible to design an ankle–foot device that will assist a walking user only during the dorsiflexion stages of the gait, avoiding interference with other stages, by using a single actuator with a simple transmission and a suspension block. This design can be improved by the use of multi-objective optimization to obtain a static set of parameters that are applicable to varying initial conditions. We present a computer simulation study of an active ankle–foot orthosis design, based on the interaction of a cam and lever with a suspension block, with the objective of assisting only with dorsiflexion during the gait cycle, leaving the rest of the movements unimpeded while reducing the complexity and weight of the device. This design is validated using a full simulation environment that includes the movements of the lower leg and foot, as they interact with our device and a ground element. As part of the design and validation, we found sets of mechanical and control parameters that provoke adequate output behavior of the orthosis to help the wearer perform a moderate-speed, normal gait. To optimize the design, we proposed three objectives to warrant ankle angle accuracy, minimal oscillations, and low energy consumption. A set of solutions was obtained with multi-objective optimization algorithms NSGA-II and RVEA to tune the parameters of the active orthosis. The solutions set from RVEA resulted in lower mean and standard deviation values for the oscillations and energy objectives in comparison to the solutions from NSGA-II, while for the MSE objective, NSGA-II obtained lower mean and standard deviation; for the energy consumption objective, the mean score using RVEA is 17% less than with NSGA-II. The orthosis is shown to be robust to differences in initial ankle angles. We observed that it is possible to obtain a broad set of solutions with a good performance during the gait cycle in controlled spaces and that in this application, the RVEA algorithm results in a better option for optimization to balance the objectives.

A network model of legal relations.

From at least the early twentieth century, legal scholars have recognized that rights and other legal relations inhere between individual legal actors, forming a vast and complex social network. Yet, no legal scholar has used the mathematical machinery of network theory to formalize these relationships. Here, we propose the first such approach by modelling a rudimentary, static set of real property relations using network theory. Then, we apply our toy model to measure the level of modularity-essentially, the community structure-among aggregations of these real property relations and associated actors. In so doing, we show that even for a very basic set of relations and actors, law may employ modular structures to manage complexity. Property, torts, contracts, intellectual property, and other areas of the law arguably reduce information costs in similar, quantifiable ways by chopping up the world of interactions between parties into manageable modules that are semi-autonomous. We also posit that our network science approach to jurisprudential issues can be adapted to quantify many other important aspects of legal systems. This article is part of the theme issue 'A complexity science approach to law and governance'.

Suspension and tyre loads estimation of an FSAE car: model development and on-track validation

In this paper a relatively simple model is presented, able to compute the loads on the suspension arms and at the tyres' contact patches of a Formula SAE car starting from telemetry data. The model is based on standard dynamic equilibrium equations, a simplified assumption for the limited-slip differential and a given front-to-rear brake distribution. The model inputs are the signals acquired with common telemetry sensors, which usually equip a Formula SAE vehicle (Inertia Measurement Unit, GPS, potentiometers, Hall sensors). In order to validate the model, some arms of the suspensions were instrumented with strain gauges, and a kinematic model of the suspension, together with its static equilibrium equation set, was set up to relate the arm loads to the forces acting at the tyre-road contact patch. The suspension kinematic and equilibrium models were first validated through static tests, with known forces applied at the wheel hub. Then, the complete vehicle model was validated by comparing the same quantities, relative to some real driving tests on a kart circuit, showing a fairly good agreement between the predicted and the measured loads. The model represents a valuable tool for vehicle design and development for the Formula SAE team.

Assisted design of data science pipelines

When designing data science (DS) pipelines, end-users can get overwhelmed by the large and growing set of available data preprocessing and modeling techniques. Intelligent discovery assistants (IDAs) and automated machine learning (AutoML) solutions aim to facilitate end-users by (semi-)automating the process. However, they are expensive to compute and yield limited applicability for a wide range of real-world use cases and application domains. This is due to (a) their need to execute thousands of pipelines to get the optimal one, (b) their limited support of DS tasks, e.g., supervised classification or regression only, and a small, static set of available data preprocessing and ML algorithms; and (c) their restriction to quantifiable evaluation processes and metrics, e.g., tenfold cross-validation using the ROC AUC score for classification. To overcome these limitations, we propose a human-in-the-loop approach for the assisteddesignofdatasciencepipelines using previously executed pipelines. Based on a user query, i.e., data and a DS task, our framework outputs a ranked list of pipeline candidates from which the user can choose to execute or modify in real time. To recommend pipelines, it first identifies relevant datasets and pipelines utilizing efficient similarity search. It then ranks the candidate pipelines using multi-objective sorting and takes user interactions into account to improve suggestions over time. In our experimental evaluation, the proposed framework significantly outperforms the state-of-the-art IDA tool and achieves similar predictive performance with state-of-the-art long-running AutoML solutions while being real-time, generic to any evaluation processes and DS tasks, and extensible to new operators.

Brain-inspired chaotic spiking backpropagation.

Spiking neural networks (SNNs) have superior energy efficiency due to their spiking signal transmission, which mimics biological nervous systems, but they are difficult to train effectively. Although surrogate gradient-based methods offer a workable solution, trained SNNs frequently fall into local minima because they are still primarily based on gradient dynamics. Inspired by the chaotic dynamics in animal brain learning, we propose a chaotic spiking backpropagation (CSBP) method that introduces a loss function to generate brain-like chaotic dynamics and further takes advantage of the ergodic and pseudo-random nature to make SNN learning effective and robust. From a computational viewpoint, we found that CSBP significantly outperforms current state-of-the-art methods on both neuromorphic data sets (e.g. DVS-CIFAR10 and DVS-Gesture) and large-scale static data sets (e.g. CIFAR100 and ImageNet) in terms of accuracy and robustness. From a theoretical viewpoint, we show that the learning process of CSBP is initially chaotic, then subject to various bifurcations and eventually converges to gradient dynamics, consistently with the observation of animal brain activity. Our work provides a superior core tool for direct SNN training and offers new insights into understanding the learning process of a biological brain.

Designing, Developing, and Validating a Set of Standardized Pictograms to Support Pediatric-Reported Gastroduodenal Symptoms

To develop and validate a set of static and animated gastroduodenal symptom pictograms for children. There were three study phases: 1: co-creation using experience design methods to develop pediatric gastroduodenal symptom pictograms (static and animated); 2: an online survey to assess acceptability, as well as face and content validity; and 3: a preference study. Phases 2 and 3 compared the novel pediatric pictograms with existing pictograms used with adult patients. Eight children aged 6-15 years (5 female) participated in phase 1, and 69 children in phase 2 (median age 13 years: IQR 9-15); an additional 49 participants were included in phase 3 (median age 15: IQR 12-17). Face and content validity were higher for the pediatric static and animated pictogram sets compared with pre-existing adult pictograms (78% vs. 78% vs. 61%). Participants with worse gastric symptoms had superior comprehension of the pediatric pictograms (χ2 (8, N=118) p< .001). All participants preferred the pediatric static pictogram set was over both the animated and adult sets (χ2 (2, N=118) p< .001). The co-creation phase resulted in the symptom concept confirmation and design of ten acceptable static and animated gastroduodenal pictograms with high face and content validity when evaluated with children aged 6 to 18. Validity was superior when children reported more problematic symptoms. Therefore, these pictograms could be used in clinical and research practice to enable standardized symptom reporting for children with gastroduodenal disorders.

“It doesn't make sense to stick with old patterns”: How leaders adapt their behavior to foster inclusion in a disruptive context

SummaryLeader behavior is essential for creating inclusive organizations. The disruptive context of the COVID‐19 pandemic forced many people to work remotely and leaders to cope with the disruption of their teams' workflows and work arrangements. However, fixed sets of leader behavior as well as stable and shared physical contexts are implicit assumptions in current knowledge and theorizing on inclusive leadership. Therefore, in this study, we first synthesize inclusive leadership literature with leader adaptability and context‐sensitive leadership studies. Next, drawing on 47 interviews with leaders and their followers, we unravel how the enactment of aspired inclusive leadership behaviors was hampered due to the pandemic‐related disruption, and explain how leaders adjusted their inclusive behaviors in response to these difficulties. From these findings, we develop a model that suggests rather than a static set of inclusive leader behaviors, inclusive leadership is enacted through the continuous adjustment on leaders' perceptions of the context and followers' feelings of inclusion.

MALWARE DETECTION : A FRAMEWORK FOR REVERSE ENGINEERED ANDROID APPLICATIONS THROUGH MACHINE LEARNING ALGORITHMS

Today, Android is one of the most used operating systems in smartphone technology. This is the main reason, Android has become the favorite target for hackers and attackers. Malicious codes are being embedded in Android applications in such a sophisticated manner that detecting and identifying an application as a malware has become the toughest job for security providers. In terms of ingenuity and cognition, Android malware has progressed to the point where they're more impervious to conventional detection techniques. Approaches based on machine learning have emerged as a much more effective way to tackle the intricacy and originality of developing Android threats. They function by first identifying current patterns of malware activity and then using this information to distinguish between identified threats and unidentified threats with unknown behavior. This research paper uses Reverse Engineered Android applications’ features and Machine Learning algorithms to find vulnerabilities present in Smartphone applications. Our contribution is twofold. Firstly, we propose a model that incorporates more innovative static feature sets with the largest current datasets of malware samples than conventional methods. Secondly, we have used ensemble learning with machine learning algorithms such as AdaBoost, SVM, etc. to improve our model's performance. Our experimental results and findings exhibit 96.24% accuracy to detect extracted malware from Android applications, with a 0.3 False Positive Rate (FPR). The proposed model incorporates ignored detrimental features such as permissions, intents, API calls, and so on, trained by feeding a solitary arbitrary feature, extracted by reverse engineering as an input to the machine.

A New Paradigm for Decision Making under Uncertainty in Signature Forensics Applications based on Neutrosophic Rule Engine

One of the most popular and legally recognized behavioral biometrics is the individual's signature, which is used for verification and identification in many different industries, including business, law, and finance. The purpose of the signature verification method is to distinguish genuine from forged signatures, a task complicated by cultural and personal variances. Analysis, comparison, and evaluation of handwriting features are performed in forensic handwriting analysis to establish whether or not the writing was produced by a known writer. In contrast to other languages, Arabic makes use of diacritics, ligatures, and overlaps that are unique to it. Due to the absence of dynamic information in the writing of Arabic signatures, it will be more difficult to attain greater verification accuracy. On the other hand, the characteristics of Arabic signatures are not very clear and are subject to a great deal of variation (features’ uncertainty). To address this issue, the suggested work offers a novel method of verifying offline Arabic signatures that employs two layers of verification, as opposed to the one level employed by prior attempts or the many classifiers based on statistical learning theory. A static set of signature features is used for layer one verification. The output of a neutrosophic logic module is used for layer two verification, with the accuracy depending on the signature characteristics used in the training dataset and on three membership functions that are unique to each signer based on the degree of truthiness, indeterminacy, and falsity of the signature features. The three memberships of the neutrosophic set are more expressive for decision-making than those of the fuzzy sets. The purpose of the developed model is to account for several kinds of uncertainty in describing Arabic signatures, including ambiguity, inconsistency, redundancy, and incompleteness. The experimental results show that the verification system works as intended and can successfully reduce the FAR and FRR.

Intelligent Target Classification Algorithm for 77G Radar Based on Correction Data Set

Traffic participant classification is crucial in autonomous driving perception. Millimeter wave radio detection and ranging radar is a cost-effective and powerful method to perform the task in adverse traffic scenarios, especially in bad inclement weather (e.g. fog, snow and rain) and poor lighting conditions. This paper presents an intelligent target classification algorithm for 77G radar based on correction data set. First, in order to handle the problem that the original data set may easily be interfered by obstacles, the angle information is filtered by analyzing the spatial information of radar signals, which means the interference clutter of obstacles can be effectively removed. Second, the primary data set is corrected using the significant difference between the micro-Doppler of the human body and car. Finally, the characteristic information of the radar signal is extracted, including distance, speed, orientation, micro-Doppler and reflection intensity, and the obtained data sets containing three types of targets (vehicles, human bodies and obstacles) are generated. The generated dynamic and static data sets are collected by sufficient experiments to construct deep learning classification models. The results show that the classification accuracy is improved by the measure of data set correction.

Learning from the 4-(dimethylamino)benzonitrile twist: Two-parameter range-separated local hybrid functional with high accuracy for triplet and charge-transfer excitations.

The recent ωLH22t range-separated local hybrid (RSLH) is shown to provide outstanding accuracy for the notorious benchmark problem of the two lowest excited-state potential energy curves for the amino group twist in 4-(dimethylamino)benzonitrile (DMABN). However, the design of ωLH22t as a general-purpose functional resulted in less convincing performance for triplet excitations, which is an important advantage of previous LHs. Furthermore, ωLH22t uses 8 empirical parameters to achieve broad accuracy. In this work, the RSLH ωLH23ct-sir is constructed with minimal empiricism by optimizing its local mixing function prefactor and range-separation parameter for only 8 excitation energies. ωLH23ct-sir maintains the excellent performance of ωLH22t for the DMABN twist and charge-transfer benchmarks but significantly improves the errors for triplet excitation energies (0.17 vs 0.24eV). Additional test calculations for the AE6BH6 thermochemistry test set and large dipole moment and static polarizability test sets confirm that the focus on excitation energies in the optimization of ωLH23ct-sir has not caused any dramatic errors for ground-state properties. Although ωLH23ct-sir cannot replace ωLH22t as a general-purpose functional, it is preferable for problems requiring a universally good description of localized and charge-transfer excitations of both singlet and triplet multiplicity. Current limitations on the application of ωLH23ct-sir and other RSLHs to the study of singlet-triplet gaps of emitters for thermally activated delayed fluorescence are discussed. This work also includes the first systematic analysis of the influence of the local mixing function prefactor and the range-separation parameter in an RSLH on different types of excitations.

MDF-DMC: A stock prediction model combining multi-view stock data features with dynamic market correlation information

Using machine learning coupled with stock price data to predict stock price trends has attracted increasing attention from data mining and machine learning communities. An accurate prediction results can help investors reduce investment risks and improve investment returns. The research on correlation stocks is one of the most important directions among many studies. Due to the high volatility and randomness of stock data, the correlation between stocks changes over time, which makes the stock correlation in static correlation stock sets often inconsistent with reality. Furthermore, various raw data related to stocks contain sufficient stock history information to analyze the future trend of stocks, but traditional prediction models cannot make good use of this information, which restricts the learning ability of the model and reduces the prediction accuracy. In this paper, we propose a stock prediction model combining multi-view stock data features with dynamic market correlation information (MDF-DMC). The model extracts stock trend features by combining multi-view raw data of a single stock with a Multi-layer Perceptron Mixer (MLP-Mixer); The improved Transformer encoder learns the correlation between the stock to be predicted and all the selected stocks in the stock market dynamically and extracts the features of the market correlation. We have conducted a large number of experiments on a total of 578 stocks in the stock markets of China and the United States, and the results show that our model has achieved excellent accuracy and returns across all data sets.