Cosine Similarity Research Articles

Optimizing Financial Decision Support Systems with Machine LearningDriven Recommendations

Introduction: The research investigates the utility of cosine similarity as an innovative recommendation system designed to assist individuals in making financial choices tailored to their unique preferences and objectives. It embarks on an extensive analysis of diverse datasets encompassing a wide array of financial products, including investment portfolios, credit card offerings, insurance plans, personal loan options, and car loan packages. Each dataset undergoes meticulous feature extraction and preprocessing to optimize the accuracy of the cosine similarity model. Method: The research then applies cosine similarity to calculate the similarity scores between individual financial products, thereby producing personalized recommendations. These recommendations are predicated on a comprehensive spectrum of input variables. The outcomes of these case studies demonstrate the potency of cosine similarity as a foundation for the development of tailored financial guidance systems. Such recommendations empower individuals to make informed decisions that are intrinsically aligned with their distinctive financial aspirations. Results: Ridge and lasso regression algorithms are deployed to develop predictive models for assessing investment preferences and evaluating potential investment returns. Conclusion: The study highlights the necessity for financial institutions and advisory platforms to invest in data quality and algorithmic sophistication to enhance the efficacy and accuracy of these financial recommendations.

Textual Patterns and Virality in X: An Analysis of Engagement in Telenovela Posts

X, previously known as Twitter, boasts 556 million active users and is widely used by businesses to engage with their audiences. In our study, we focused on TV Globo's telenovela "Terra e Paixão" broadcast in 2023, to analyze the impact of textual patterns on post virality using natural language processing techniques. Techniques like sentiment analysis, Part-Of-Speech Tagging, reinforcement scoring, TF-IDF, semantic similarity, and cosine similarity were utilized to identify attributes that contribute to a post's success, aiming to enhance marketing strategies. We employed language models like BERT, RoBERTa, and e5 in our analysis. Our findings indicate that while various metrics affect post engagement, the challenge remains complex. Textual characteristics, although essential, do not fully explain a publication's popularity, underscoring the need for a multifaceted approach to understanding social media dynamics.

QC-GN2oMS2: a Graph Neural Net for High Resolution Mass Spectra Prediction.

Predicting the mass spectrum of a molecular ion is often accomplished via three generalized approaches: rules-based methods for bond breaking, deep learning, or quantum chemical (QC) modeling. Rules-based approaches are often limited by the conditions for different chemical subspaces and perform poorly under chemical regimes with few defined rules. QC modeling is theoretically robust but requires significant amounts of computational time to produce a spectrum for a given target. Among deep learning techniques, graph neural networks (GNNs) have performed better than previous work with fingerprint-based neural networks in mass spectra prediction. To explore this technique further, we investigate the effects of including quantum chemically derived information as edge features in the GNN to increase predictive accuracy. The models we investigated include categorical bond order, bond force constants derived from extended tight-binding (xTB) quantum chemistry, and acyclic bond dissociation energies. We evaluated these models against a control GNN with no edge features in the input graphs. Bond dissociation enthalpies yielded the best improvement with a cosine similarity score of 0.462 relative to the baseline model (0.437). In this work we also apply dynamic graph attention which improves performance on benchmark problems and supports the inclusion of edge features. Between implementations, we investigate the nature of the molecular embedding for spectra prediction and discuss the recognition of fragment topographies in distinct chemistries for further development in tandem mass spectrometry prediction.

DeepCKID: A Multi-Head Attention-Based Deep Neural Network Model Leveraging Classwise Knowledge to Handle Imbalanced Textual Data

This paper presents DeepCKID, a Multi-Head Attention (MHA)-based deep learning model that exploits statistical and semantic knowledge corresponding to documents across different classes in the datasets to improve the model’s ability to detect minority class instances in imbalanced text classification. In this process, corresponding to each document, DeepCKID extracts — (i) word-level statistical and semantic knowledge, namely, class correlation and class similarity corresponding to each word, based on its association with different classes in the dataset, and (ii) class-level knowledge from the document using n-grams and relation triplets corresponding to classwise keywords present, identified using cosine similarity utilizing Transformers-based Pre-trained Language Models (PLMs). DeepCKID encodes the word-level and class-level features using deep convolutional networks, which can learn meaningful patterns from them. At first, DeepCKID combines the semantically meaningful Sentence-BERT document embeddings and word-level feature matrix to give the final document representation, which it further fuses to the different classwise encoded representations to strengthen feature propagation. DeepCKID then passes the encoded document representation and its different classwise representations through an MHA layer to identify the important features at different positions of the feature subspaces, resulting in a latent dense vector accentuating its association with a particular class. Finally, DeepCKID passes the latent vector to the softmax layer to learn the corresponding class label. We evaluate DeepCKID over six publicly available Amazon reviews datasets using four Transformers-based PLMs. We compare DeepCKID with three approaches and four ablation-like baselines. Our study suggests that in most cases, DeepCKID outperforms all the comparison approaches, including baselines.

Novel Fault Protection Method for Flexible DC Power Systems

A fault protection method is proposed for flexible direct-current (DC) power systems based on the cosine similarity of currents in current-limiting reactors. The typical characteristics of external and internal faults in a flexible DC power system were analyzed. The principles of cosine similarity and the fault current characteristics of current-limiting reactors were used to distinguish between internal and external faults. The ratio of the average positive and negative voltages of the current-limiting reactor was then used to distinguish the fault types and fault line. Finally, a simulation model of a circular flexible DC power system was constructed using MATLAB/Simulink software (2018b). The simulation results show that the proposed protection scheme can quickly and accurately identify the location and type of faults, and has a strong ability to withstand fault resistance and is less affected by distributed capacitance, noise, and communication delay.

Fine-Grained Few-Shot Image Classification Based on Feature Dual Reconstruction

Fine-grained few-shot image classification is a popular research area in deep learning. The main goal is to identify subcategories within a broader category using a limited number of samples. The challenge stems from the high intra-class variability and low inter-class variability of fine-grained images, which often hamper classification performance. To overcome this, we propose a fine-grained few-shot image classification algorithm based on bidirectional feature reconstruction. This algorithm introduces a Mixed Residual Attention Block (MRA Block), combining channel attention and window-based self-attention to capture local details in images. Additionally, the Dual Reconstruction Feature Fusion (DRFF) module is designed to enhance the model’s adaptability to both inter-class and intra-class variations by integrating features of different scales across layers. Cosine similarity networks are employed for similarity measurement, enabling precise predictions. The experiments demonstrate that the proposed method achieves classification accuracies of 96.99%, 98.53%, and 89.78% on the CUB-200-2011, Stanford Cars, and Stanford Dogs datasets, respectively, confirming the method’s efficacy in fine-grained classification tasks.

Deep probabilistic direction prediction in 3D with applications to directional dark matter detectors

We present the first method to probabilistically predict 3D direction in a deep neural network model. The probabilistic predictions are modeled as a heteroscedastic von Mises-Fisher distribution on the sphere S2 , giving a simple way to quantify aleatoric uncertainty. This approach generalizes the cosine distance loss which is a special case of our loss function when the uncertainty is assumed to be uniform across samples. We develop approximations required to make the likelihood function and gradient calculations stable. The method is applied to the task of predicting the 3D directions of electrons, the most complex signal in a class of experimental particle physics detectors designed to demonstrate the particle nature of dark matter and study solar neutrinos. Using simulated Monte Carlo data, the initial direction of recoiling electrons is inferred from their tortuous trajectories, as captured by the 3D detectors. For 40 keV electrons in a 70% He 30% CO2 gas mixture at STP, the new approach achieves a mean cosine distance of 0.104 (26∘) compared to 0.556 (64∘) achieved by a non-machine learning algorithm. We show that the model is well-calibrated and accuracy can be increased further by removing samples with high predicted uncertainty. This advancement in probabilistic 3D directional learning could increase the sensitivity of directional dark matter detectors.

Performance of GPT-4 on Chinese Nursing Examination: Potentials for AI-Assisted Nursing Education Using Large Language Models.

The performance of GPT-4 in nursing examinations within the Chinese context has not yet been thoroughly evaluated. To assess the performance of GPT-4 on multiple-choice and open-ended questions derived from nursing examinations in the Chinese context. The data sets of the Chinese National Nursing Licensure Examination spanning 2021 to 2023 were used to evaluate the accuracy of GPT-4 in multiple-choice questions. The performance of GPT-4 on open-ended questions was examined using 18 case-based questions. For multiple-choice questions, GPT-4 achieved an accuracy of 71.0% (511/720). For open-ended questions, the responses were evaluated for cosine similarity, logical consistency, and information quality, all of which were found to be at a moderate level. GPT-4 performed well at addressing queries on basic knowledge. However, it has notable limitations in answering open-ended questions. Nursing educators should weigh the benefits and challenges of GPT-4 for integration into nursing education.

Improved research on coral bleaching detection model based on FCOS model

Coral bleaching detection is vital for assessing coral reef health. This paper introduces FCOS_EfficientNET, an improved model that enhances accuracy, recall, and real-time performance in coral bleaching detection. Utilizing EfficientNet as the backbone, the model optimizes parameter throughput. We adopt the ReLU activation function and utilize cosine similarity and softmax to assign weights to datasets, modifying the attention structure to reduce memory consumption. The model also integrates BiFPN for better feature extraction and employs an improved training method to enhance detection accuracy. To cater to different scenarios, we have developed four variants: FCOS_EfficientNETb0, FCOS_EfficientNETb1, FCOS_EfficientNETb2, and FCOS_EfficientNETb3. Experimental results on the MS COCO dataset show that FCOS_EfficientNETb3 achieves a mean average precision (mAP) of 48.5%, while FCOS_EfficientNETb0 reaches a frame rate of 167.17 fps, highlighting the superior performance of the series. On a custom coral bleaching detection dataset, FCOS_EfficientNETb3 achieves 81.5% accuracy and a 59.3% recall rate, demonstrating the effectiveness of the model. FCOS_EfficientNETb1 and FCOS_EfficientNETb2 offer a balance between operations per second, frame rate, and mAP, making them suitable for mobile and edge computing. These models effectively track movement or changes in marine traffic around coral reefs with moderate fps and recall rates.

Prototype as query for few shot semantic segmentation

Few-shot Semantic Segmentation (FSS) was proposed to segment unseen classes in a query image, referring to only a few annotated examples named support images. One of the characteristics of FSS is spatial inconsistency between query and support targets, e.g., texture or appearance. This greatly challenges the generalization ability of methods for FSS, which requires to effectively exploit the dependency of the query image and the support examples. Most existing methods abstracted support features into prototype vectors and implemented the interaction with query features using cosine similarity or feature concatenation. However, this simple interaction may not capture spatial details in query features. To address this limitation, some methods utilized pixel-level support information by computing pixel-level correlations between paired query and support features implemented with the attention mechanism of Transformer. Nevertheless, these approaches suffer from heavy computation due to dot-product attention between all pixels of support and query features. In this paper, we propose a novel framework, termed ProtoFormer, built upon the Transformer architecture, to fully capture spatial details in query features. ProtoFormer treats the abstracted prototype of the target class in support features as the Query and the query features as Key and Value embeddings, which are input to the Transformer decoder. This approach enables better capture of spatial details and focuses on the semantic features of the target class in the query image. The output of the Transformer-based module can be interpreted as semantic-aware dynamic kernels that filter the segmentation mask from the enriched query features. Extensive experiments conducted on PASCAL-5i\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$5^{i}$$\\end{document} and COCO-20i\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$20^{i}$$\\end{document} datasets demonstrate that ProtoFormer significantly outperforms the state-of-the-art methods in FSS.

Synthetic-Aperture Radar Radio-Frequency Interference Suppression Based on Regularized Optimization Feature Decomposition Network

Synthetic-aperture radar (SAR) can work in all weather conditions and at all times, and satellite-borne radar has the characteristics of short revisiting period and large imaging width. Therefore, satellite-borne synthetic-aperture radar has been widely deployed, and the SAR images have been widely used in geographic mapping, radar interpretation, ship detection, and other fields. Satellite-borne synthetic-aperture radar is also susceptible to various types of intentional or unintentional interference during the imaging process, and because the interference is a direct wave, its power is much stronger than the wave reflected by targets. As a common interference pattern, radio-frequency interference widely exists in various satellite-borne synthetic-aperture radars, which seriously deteriorates SAR image quality. In order to solve the above problems, this paper proposes a feature decomposition network to suppress interference based on regularization optimization. The contributions of this work are as follows: 1. By analyzing the performance limitations of the existing methods, this work proposes a novel regularization method for radio-frequency interference suppression tasks. From the perspective of data distribution histograms and residual components, the proposed method eliminates the variable components introduced by common regularization, greatly reduces the difficulty of data mapping, and significantly improves its robustness and performance. 2. This work proposes a feature decomposition network, where the feature decomposition module contains two parts; one part only represents the interference signal, and the other part only represents the radar signal. The neurons representing the interference signal are discarded, and the neurons representing the radar signal are used as input for the subsequent network. A cosine similarity constraint is used to separate the interference from the network as much as possible. Finally, this method is validated on the MiniSAR dataset and Sentinel-1A dataset.

Fed4UL: A Cloud–Edge–End Collaborative Federated Learning Framework for Addressing the Non-IID Data Issue in UAV Logistics

Artificial intelligence and the Internet of Things (IoT) have brought great convenience to people’s everyday lives. With the emergence of edge computing, IoT devices such as unmanned aerial vehicles (UAVs) can process data instantly at the point of generation, which significantly decreases the requirement for on-board processing power and minimises the data transfer time to enable real-time applications. Meanwhile, with federated learning (FL), UAVs can enhance their intelligent decision-making capabilities by learning from other UAVs without directly accessing their data. This facilitates rapid model iteration and improvement while safeguarding data privacy. However, in many UAV applications such as UAV logistics, different UAVs may perform different tasks and cover different areas, which can result in heterogeneous data and add to the problem of non-independent and identically distributed (Non-IID) data for model training. To address such a problem, we introduce a novel cloud–edge–end collaborative FL framework, which organises and combines local clients through clustering and aggregation. By employing the cosine similarity, we identified and integrated the most appropriate local model into the global model, which can effectively address the issue of Non-IID data in UAV logistics. The experimental results showed that our approach outperformed traditional FL algorithms on two real-world datasets, CIFAR-10 and MNIST.

Integrating muscle fiber orientation from visible human data into radiotherapy target volumes.

Objective.A major challenge in treatment of tumors near skeletal muscle is defining the target volume for suspected tumor invasion into the muscle. This study develops a framework that generates radiation target volumes with muscle fiber orientation directly integrated into their definition. The framework is applied to nineteen sacral tumor patients with suspected infiltration into surrounding muscles.Approach.To compensate for the poor soft-tissue contrast of CT images, muscle fiber orientation is derived from cryo-images of two cadavers from the human visible project (VHP). The approach consists of (a) detecting image gradients in the cadaver images representative of muscle fibers, (b) mapping this information onto the patient image, and (c) embedding the muscle fiber orientation into an expansion method to generate patient-specific clinical target volumes (CTV). The validation tested the consistency of image gradient orientation across VHP subjects for the piriformis, gluteus maximus, paraspinal, gluteus medius, and gluteus minimus muscles. The model robustness was analyzed by comparing CTVs generated using different VHP subjects. The difference in shape between the new CTVs and standard CTV was analyzed for clinical impact.Main results.Good agreement was found between the image gradient orientation across VHP subjects, as the voxel-wise median cosine similarity was at least 0.86 (for the gluteus minimus) and up to 0.98 for the piriformis. The volume and surface similarity between the CTVs generating from different VHP subjects was on average at least 0.95 and 5.13 mm for the Dice Similarity Coefficient and the Hausdorff 95% Percentile Index, showing excellent robustness. Finally, compared to the standard CTV with different margins in muscle and non-muscle tissue, the new CTV margins are reduced in muscle tissue depending on the chosen clinical margins.Significance.This study implements a method to integrate muscle fiber orientation into the target volume without the need for additional imaging.

Topological consistency with low imperceptibility for graph adversarial attacks

Recent research shows that graph neural networks (GNNs) are easy to receive disruptions due to the lack of robustness, the phenomenon that poses a serious security threat. Currently, most efforts to attack GNNs mainly use gradient information to guide the attacks. However, the unreliability of gradient information, and the perceptibility of adversarial examples pose challenges that impede further progress in researching on graph adversarial attacks. From the unreliability of gradient information, we propose a Graph Distance Topological Consistency (GDTC). The scheme introduces graph connectivity, geodesic distance, cosine similarity, and Minkowski distance to construct the similarity matrices of the input space and embedding space of the surrogate model. The difference between the two similarities matrices is constrained during the training process of the surrogate model so that the surrogate model fully learns the topology of the original graph. From adversarial examples perceptibility, we propose attack loss with homogeneity restriction. Experiments show that GDTC can study the topological information of the original graph, enhance the reliability of gradient information, and significantly boost attack performance.

Unveiling Hidden Insights in Gas Chromatography Data Analysis with Generative Adversarial Networks

The gas chromatography analysis method for chemical substances enables accurate analysis to precisely distinguish the components of a mixture. This paper presents a technique for augmenting time-series data of chemicals measured by gas chromatography instruments with artificial intelligence techniques such as generative adversarial networks (GAN). We propose a novel GAN algorithm called GCGAN for gas chromatography data, a unified model of autoencoder (AE) and GAN for effective time-series data learning with an attention mechanism. The proposed GCGAN utilizes AE to learn a limited number of data more effectively. We also build a layer of high-performance generative adversarial neural networks based on the analysis of the features of data measured by gas chromatography instruments. Then, based on the proposed learning, we synthesize the features embedded in the gas chromatography data into a feature distribution that extracts the temporal variability. GCGAN synthesizes the features embedded in the gas chromatography data into a feature distribution that extracts the temporal variability of the data over time. We have fully implemented the proposed GCGAN and experimentally verified that the data augmented by the GCGAN have the characteristic properties of the original gas chromatography data. The augmented data demonstrate high quality with the Pearson correlation coefficient, Spearman correlation coefficient, and cosine similarity all exceeding 0.9, significantly enhancing the performance of AI classification models by 40%. This research can be effectively applied to various small dataset domains other than gas chromatography data, where data samples are limited and difficult to obtain.

Evaluating the performance of multilingual models in answer extraction and question generation

Multiple-choice test generation is one of the most complex NLP problems, especially in languages other than English, where there is a lack of prior research. After a review of the literature, it has been verified that some methods like the usage of rule-based systems or primitive neural networks have led to the application of a recent architecture, the Transformer architecture, in the tasks of Answer Extraction (AE) and Question Generation (QG). Thereby, this study is centred in searching and developing better models for the AE and QG tasks in Spanish, using an answer-aware methodology. For this purpose, three multilingual models (mT5-base, mT0-base and BLOOMZ-560 M) have been fine-tuned using three different datasets: a translation to Spanish of the SQuAD dataset; SQAC, which is a dataset in Spanish; and their union (SQuAD + SQAC), which shows slightly better results. Regarding the models, the performance of mT5-base has been compared with that found in two newer models, mT0-base and BLOOMZ-560 M. These models were fine-tuned for multiple tasks in literature, including AE and QG, but, in general, the best results are obtained from the mT5 models trained in our study with the SQuAD + SQAC dataset. Nonetheless, some other good results are obtained from mT5 models trained only with the SQAC dataset. For their evaluation, the widely used BLEU1-4, METEOR and ROUGE-L metrics have been obtained, where mT5 outperforms some similar research works. Besides, CIDEr, SARI, GLEU, WER and the cosine similarity metrics have been calculated to present a benchmark within the AE and QG problems for future work.

Autonomous and reliable fingerprint map maintenance for indoor positioning system

In WiFi-based crowdsourced fingerprinting indoor positioning systems (IPS), the signal space of WiFi APs changes over time. To autonomously maintain a highly reliable localization accuracy, the fingerprint database needs regular maintenance and updates. However, this process is labor-intensive and time-consuming due to the appearance of faulty access points (APs) in crowdsourced samples received from target clients, which are mainly used for autonomous maintenance and updates. Additionally, if some APs are relocated, removed, or replaced in the indoor environment after fingerprint map construction, it can severely impact IPS accuracy. In this paper, we present Dynamic Online Fingerprint Learning with Altered APs status-aware and updating (DOFLAAU) for IPS. Our systematic methodology involves several key steps. First, we employ the Modified Group Method of Data Handling (GMDH) neural network regression for autonomous AP status monitoring and alteration detection. Next, we perform autonomous APs subset sample selection using a modified Thomson model, followed by outlier feature detection and removal based on the weighted sample density rate. Thirdly, we propose a dynamic online fingerprint map updating model using the cosine similarity distance metric. Lastly, we adopted an effective signal distance-based weighted k-nearest neighbor algorithm to improve IPS accuracy. The experiment results validate the efficiency and effectiveness of our proposed algorithms in terms of better localization accuracy compared to peer algorithms.

IC points weight learning-based GCN and improving feature distribution for industrial fault diagnosis

Industrial fault diagnosis (FD) is crucial to detect the causes of faults in a timely manner and solve the problems to maintain stable production. Adequately considering the spatial structure of the samples and their features is an urgent problem for better FD. This paper focused on the points located in the within-class boundaries and between-class intersection region (IC points) and proposed IC point weight learning-based graph convolution network (GCN) and improving feature distribution (ICWGCN-FD) method. Firstly, ICWGCN-FD assigns appropriate weights to IC points specially allocated by graph learning layer. Secondly, a cosine distance loss term is particularly designed to make the IC points closer to the center of their respective classes and further away from their nearest neighbor points belongs to other classes. Finally, the spatial feature distribution of the overall samples is improved under the influence of the message passing function possessed by GCN, thus presenting a clearer distribution. Two cases from industrial simulation are carried out to validate the practical feasibility and superiority of ICWGCN-FD, compared with other classic and recent graph neural networks. In addition, the 3D visualization results correspond to the principles of the method, and an ablation experiment successfully verified the validity of each component.

P-685 An artificial intelligence based medication regimen recommendation model for optimizing the treatment outcomes of controlled ovarian stimulation

Abstract Study question Can a machine learning (ML) algorithm suggest an optimal controlled ovarian stimulation (COS) protocol for maximizing the number of retrieved and mature (MII) oocytes? Summary answer Based on historical electronic medical record patient data, using ML algorithm based medication regimen recommendation model may increase the number of retrieved oocytes. What is known already The application of artificial intelligence in the field of reproductive medicine is mainly focused on the image analysis of gamete and embryo, and prediction of treatment outcomes. The research on the optimization of COS process is limited. Previous studies mainly focused on one or several types of COS protocols such as GnRH-a protocol and GnRH-A protocol. And most of them aimed to assist single-node decision-making during COS process such as optimal gonadotropins starting dose and trigger day. Study design, size, duration The data used for developing this model consists of 38,355 COS cycles from women who underwent their first IVF/ICSI treatment in a large fertility center between January 2008 to December 2022, of which 33,853 were used as the model training set and 4,502 were used as the test set. This dataset contains a subset of 8,382 cycles with information about MII oocytes. Participants/materials, setting, methods The goal of model was to recommend the optimal COS protocol based on optimizing COS outcomes. The similarity of patient’s characteristic was used as model input. XGBoost algorithm was constructed to calculate the similarity of COS outcomes. The accuracy was compared with the patient similarity calculation method of KNN+Euclidean distance and KNN+cosine distance using mean absolute error (MAE). The performance of different models in improving the average number of retrieved and MII oocytes was compared. Main results and the role of chance The patient similarity calculation method based on XGBoost had lowest MAE when predicting the number of retrieved oocytes (XGBoost vs KNN+Euclidean distance+k=500 vs KNN+Euclidean distance+k=1000 vs KNN+cosine distance+k=500 vs KNN+cosine distance+k=1000: 3.729 vs 3.810 vs 3.815 vs 3.785 vs 3.788). Including AMH could improve the accuracy of all recommendation models in predicting protocols and the number of retrieved oocytes. When the optimal number of MII oocytes is used as the protocol recommended target, the patient similarity calculation method based on XGBoost had lowest MAE (XGBoost vs KNN+Euclidean distance+k=500 vs KNN+Euclidean distance+k=1000 vs KNN+cosine distance+k=500 vs KNN+cosine distance+k=1000: 2.868 vs 3.397 vs 3.452 vs 3.356 vs 3.395). The COS medication regimen recommended by the patient similarity calculation method based on COS treatment outcomes has resulted in higher numbers of retrieved oocytes in patients compared to those who did not adopt recommendation (P < 0.01). The medication regimen recommended model based on KNN+Euclidean distance algorithm (k = 500) has the best performance in improving the average number of retrieved oocytes（adopt recommendation vs not adopt recommendation: 13.01±6.61 vs 10.09±6.67, P < 0.01）There was no statistical difference in the number of MII oocytes between patients receiving the recommendation and those not adopt recommendation. Limitations, reasons for caution The medication regimen recommendation model is based solely on the predicted number of retrieved and MII oocytes, not including adverse treatment outcomes such as ovarian hyperstimulation syndrome. Moreover, the database used to train and validate the model is from a single fertility center. Wider implications of the findings The recommendation model is helpful to improve the number of retrieved oocytes. Based on the recommended COS protocol, further mining the medication content, medication time distribution and medication sequence transformation from the similar patient set can assist clinicians to develop individualized COS medication regimen for infertile patients. Trial registration number not applicable

Designing unique and high-performance Al alloys via machine learning: Mitigating data bias through active learning

Data-driven modelling, such as machine learning (ML), has great potential to streamline the complexity involved in designing new alloys. However, such powerful predictive models require a high-quality dataset, which is limited in the alloy field due to selection and reporting biases. These biases lead to out-of-distribution (OOD) regions where ML models suffer from predictive performance degradation, limiting the design of innovative alloys. To overcome this problem, we propose a ML-based design strategy for unique and high-performance aluminium (Al) alloys, incorporating multi-objective genetic algorithm (MOGA) and active learning. To guide active learning, the cosine similarity metric embedded MOGA pinpoints unique and high-strength Al alloys in OOD regions for experiments. Our study demonstrates the deficiency of initial ML models when trained on the biased dataset and subsequent improvement in retrained models after applying active learning with alloys suggested by MOGA. On this basis, a new Al alloy that is distinct from the existing dataset is developed with a yield tensile strength of 688 MPa, ultimate tensile strength of 738 MPa, and elongation of 7.5 %. This finding highlights the importance of the inclusion of OOD results and the efficiency of ML in alloy design.