Combining Machine Learning Research Articles

Construction of an Intelligent Identification Model for Drugs in Near Infrared Spectroscopy and Research on Drog Classification based on Improved Deep Algorithm

Near-infrared spectroscopy has important applications in drug and food identification. Combining machine learning with near-infrared spectroscopy to achieve intelligent identification of drugs has become a research hotspot in recent years. To solve the problem of machine learning’s inefficiency in classifying small-scale data, a drug identification model based on near-infrared spectroscopy combined with a random fading depth belief network is proposed. Aiming at the problem that the training time of the machine learning algorithm is too long, the extreme learning machine is used to replace the back propagation algorithm to optimize the stack sparse auto-encoder network. Additionally, the stack sparse auto-encoder algorithm based on extreme learning machine algorithm is constructed. The study found that the precision of the Dropout Deep Belief Network model was 99.12%, which was higher than the other three models. Additionally, the area under the curve value of the Dropout Deep Belief Network model was 0.87, which was 0.04 higher than the binary whale optimization algorithm model, 0.26 higher than the factor decomposition machine and depth neural network model, and 0.05 higher than the random forest network model. The sparse auto-encoder algorithm based on the extreme learning machine algorithm achieved a precision of 99.72%. The study proposes two algorithm models that can effectively identify drugs using near-infrared spectroscopy. This has a positive impact on the medical industry and the safety of patients' lives and health.

Mnemonic politics among Philippine voters: a social media measurement approach

ABSTRACT Nostalgic rhetoric has proven effectiveness among frustrated voters in democratic elections. Studies have introduced various measures for nostalgic sentiment, but the debate remains with regard to finding a suitable instrument. Using a social media approach, we propose a method to identify nostalgic rhetoric and investigate observable patterns, leveraging the Philippine presidential election in May 2022. We conduct text-as-data analysis and investigate sub-national and individual variation in nostalgia for Ferdinand Marcos Sr. Our approach, which combines machine learning and human labelling, identifies a significant share of tweets about Marcos Jr. as having nostalgia for Marcos Sr. Using these nostalgic tweets as a measure for authoritarian nostalgia, we explore regional variation in Marcos nostalgia across Philippines’ provinces and run statistical analysis with existing predictors for Marcos nostalgia. Our analysis shows that voters in provinces with higher support for Rodrigo Duterte in the 2016 election are more likely to post nostalgic tweets. We further explore individual variation and find that individual accounts that express negative assessment of their current status are more likely to use nostalgic rhetoric. Combined results from this paper provide among the first empirical analysis of large-scale data on how voters discuss and distribute nostalgic rhetoric for a former dictatorship.

Forecast urban ecosystem services to track climate change: Combining machine learning and emergy spatial analysis

It is important to simulate urban ecosystem services and provide guidance for their conservation and management under complex scenarios. We developed software for the rapid computation of ecosystem services based on emergy spatial analysis, machine learning, and the DLUCP model to predict the precipitation based on climate change and land use changes in five Chinese cities. The LSTM model is used to predict monthly precipitation for the cities based on historical data of 720 months. We also utilized the land use change prediction model based on the idea of distribution to improve the accuracy of future land use changes in China. The results show that based on the predicted values in 2025, an increase in ES is observed for Beijing, Wuhan, and Chongqing in 2020–2025 under both SSP1-RCP2.6 and SSP5-RCP8.5 scenarios, while the ES in Guangzhou and Shenzhen decreased. Furthermore, among the three types of ecosystems, the woodland ecosystem contributed the most to the total ES. The change in the woodland area had the greatest impact. Therefore, it is recommended that ecosystem conservation and restoration should focus on woodland ecosystem, and proactively address climate change to help achieve carbon peak and neutrality goals. The accuracy and applicability of the software are also tested.

Predicting accurate batch queue wait times on production supercomputers by combining machine learning techniques

AbstractThe ability to accurately predict when a job on a supercomputer will leave the queue and start to run is not only beneficial for providing insights to users, but can also help enable non‐traditional HPC workloads that are not necessarily suited to the batch queue style‐approach that is ubiquitous on production HPC machines. However there are numerous challenges in achieving such a prediction with high accuracy, not least because the queue's state can change rapidly and depend upon many factors. In this work, we explore a novel machine learning approach for predicting queue wait times, hypothesising that such a model can capture the complex behavior resulting from the queue policy and other interactions to generate accurate job start times. For ARCHER2 (HPE Cray EX), Cirrus (HPE 8600), and 4‐cabinet (HPE Cray EX) we explore how different machine learning approaches and techniques improve the accuracy of our predictions, comparing against the estimation generated by Slurm. By combining categorization and regression models, we demonstrate that our approach delivers the most accurate predictions across our machines of interest, with the result of this work being the ability to predict job start times within 1 min of the actual start time for around 65% of jobs on ARCHER2 and 4‐cabinet, and 76% of jobs on Cirrus. When compared against what Slurm can deliver, via the backfill plugin, this represents around 3.8 times better accuracy on ARCHER2 and 18 times better for Cirrus. Furthermore our approach can accurately predicting the start time for three quarters of all job within 10 min of the actual start time on ARCHER2 and 4‐cabinet, and for 90% of jobs on Cirrus. Whilst the initial driver of this work was to better facilitate non‐traditional, interactive and urgent, workloads on HPC machines, the insights gained can also be used to provide wider benefits to users, enrich existing batch queue systems, and inform supercomputing center policy also.

Does wealth equate to happiness? an 11-year panel data analysis exploring socio-economic indicators and social media metrics.

The Easterlin paradox questions the link between economic growth and national well-being, emphasizing the necessity to explore the impact of economic elasticity, income inequality, and their temporal and spatial heterogeneity on subjective happiness. Despite the importance of these factors, few studies have examined them together, thus ongoing debates about the impact of economics on well-being persist. To fill this gap, our analysis utilizes 11 years of panel data from 31 provinces in China, integrating macroeconomic indicators and social media content to reassess the Easterlin paradox. We use GDP per capita and the Gini coefficient as proxies for economic growth and income inequality, respectively, to study their effects on the subjective well-being expressed by citizens on social media in mainland China. Our approach combines machine learning and fixed effects models to evaluate these relationships. Key findings include: (1) In temporal relationships, a 46.70% increase in GDP per capita implies a 0.38 increase in subjective well-being, while a 0.09 increase in the Gini coefficient means a 1.47 decrease in subjective well-being. (2) In spatial relationships, for every 46.70% increase in GDP per capita, subjective well-being rises by 0.51; however, this relationship is buffered by unfair distribution, and GDP per capita no longer significantly affects subjective well-being when the Gini index exceeds 0.609. This study makes a synthetic contribution to the debate on the Easterlin paradox, indicating that economic growth can enhance well-being if income inequality is kept below a certain level. Although these results are theoretically enlightening for the relationship between economics and national well-being globally, this study's sample comes from mainland China. Due to differences in cultural, economic, and political factors, further research is suggested to explore these dynamics globally.

Boosting Clear Cell Renal Carcinoma-Specific Drug Discovery Using a Deep Learning Algorithm and Single-Cell Analysis.

Clear cell renal carcinoma (ccRCC), the most common subtype of renal cell carcinoma, has the high heterogeneity of a highly complex tumor microenvironment. Existing clinical intervention strategies, such as target therapy and immunotherapy, have failed to achieve good therapeutic effects. In this article, single-cell transcriptome sequencing (scRNA-seq) data from six patients downloaded from the GEO database were adopted to describe the tumor microenvironment (TME) of ccRCC, including its T cells, tumor-associated macrophages (TAMs), endothelial cells (ECs), and cancer-associated fibroblasts (CAFs). Based on the differential typing of the TME, we identified tumor cell-specific regulatory programs that are mediated by three key transcription factors (TFs), whilst the TF EPAS1/HIF-2α was identified via drug virtual screening through our analysis of ccRCC's protein structure. Then, a combined deep graph neural network and machine learning algorithm were used to select anti-ccRCC compounds from bioactive compound libraries, including the FDA-approved drug library, natural product library, and human endogenous metabolite compound library. Finally, five compounds were obtained, including two FDA-approved drugs (flufenamic acid and fludarabine), one endogenous metabolite, one immunology/inflammation-related compound, and one inhibitor of DNA methyltransferase (N4-methylcytidine, a cytosine nucleoside analogue that, like zebularine, has the mechanism of inhibiting DNA methyltransferase). Based on the tumor microenvironment characteristics of ccRCC, five ccRCC-specific compounds were identified, which would give direction of the clinical treatment for ccRCC patients.

Challenges and Outcomes of Combining Machine Learning with Software-Defined Networking for Network Security and management Purpose: A Review

Current research in data dissemination in Vehicular Ad Hoc Networks (VANETs) has examined different approaches and frameworks to enhance the effectiveness and dependability of information sharing between vehicles on the road. The integration of Machine Learning (ML) with Software-Defined Networking (SDN) has fundamentally transformed the field of network administration and security. This paper specifically addresses the challenges faced by traditional network architectures in effectively handling the increasing amount of data and complex applications. Software-Defined Networking (SDN), a cutting-edge framework, separates the control of network operations from the actual forwarding of data, offering a versatile and cost-effective solution. The combination of Software-Defined Networking (SDN) and Machine Learning (ML) allows for the extraction of valuable information from network data, leading to enhanced network management and the facilitation of predictive analytics. The aim of this study is to examine the feasibility and challenges of incorporating machine learning into software-defined networking (SDN), focusing particularly on practical applications. Integrating Machine Learning (ML) into Software-Defined Networking (SDN) presents challenges, including the requirement for robust algorithms to detect patterns and ensure security. It is crucial to carry out the tasks of developing and implementing machine learning models for real-time predictions and ensuring the robustness of the system. Research is essential to strike a balance between the transformative abilities of ML-SDN and the practical network environments. This helps to improve the resilience, security, and adaptability of network infrastructures in the digital age.

Timing along the cardiac cycle modulates neural signals of reward-based learning

Natural fluctuations in cardiac activity modulate brain activity associated with sensory stimuli, as well as perceptual decisions about low magnitude, near-threshold stimuli. However, little is known about the relationship between fluctuations in heart activity and other internal representations. Here we investigate whether the cardiac cycle relates to learning-related internal representations – absolute and signed prediction errors. We combined machine learning techniques with electroencephalography with both simple, direct indices of task performance and computational model-derived indices of learning. Our results demonstrate that just as people are more sensitive to low magnitude, near-threshold sensory stimuli in certain cardiac phases, so are they more sensitive to low magnitude absolute prediction errors in the same cycles. However, this occurs even when the low magnitude prediction errors are associated with clearly suprathreshold sensory events. In addition, participants exhibiting stronger differences in their prediction error representations between cardiac cycles exhibited higher learning rates and greater task accuracy.

Machine learning and IoT Based Robotic Solution for Home Wrist Rehabilitation through Motion Capture Enabled Architecture

The present study investigates the effectiveness of an integrated approach that combines machine learning, robotic rehabilitation, and traditional methods for wrist rehabilitation among the hemiparesis population. Subjects with robotic rehabilitation at the four-week intervention period showed the ball’s significant difference in the flexion and extension angles and demonstrated the better overall outcomes from the robotic intervention . The machine learning models, in particular, the artificial neural networks, provided the prediction of rehabilitation outcomes and demonstrated the high accuracy, high precision, high recall, and high F1 score. This study supports the idea of the beneficial effects of combining robotic rehabilitation with machine learning algorithms and supports the notion that the two benefits could bring high synergistic effects above their individual potentials. The report also contributes to the development of the rehabilitation science of wrists, given that few studies have explored the applications of artificial intelligence methods or robotic rehabilitation beyond the context of feasibility reports. The findings of the current research could contribute useful insights for the decision-making practices of therapists and healthcare providers, also given the reliance on large data contributed by the previous interventions. Future study could investigate the long-term results, the cost-effectiveness of the approaches, and assist in developing the potential results of customization for individual patients.

More Efficient Manual Review of Automatically Transcribed Tabular Data

Any machine learning method for transcribing historical text requires manual verification and correction, which is often time-consuming and expensive. Our aim is to make it more efficient. Previously, we developed a machine learning model to transcribe 2.3 million handwritten occupation codes from the Norwegian 1950 census. Here, we manually review the 90,000 codes (3%) for which our model had the lowest confidence scores. We allocated these codes to human reviewers, who used our custom annotation tool to review them. The reviewers agreed with the model's labels 31.9% of the time. They corrected 62.8% of the labels, and 5.1% of the images were uncertain or assigned invalid labels. 9,000 images were reviewed by multiple reviewers, resulting in an agreement of 86.4% and a disagreement of 9%. The results suggest that one reviewer per image is sufficient. We recommend that reviewers indicate any uncertainty about the label they assign to an image by adding a flag to their label. Our interviews show that the reviewers performed internal quality control and found our custom tool to be useful and easy to operate. We provide guidelines for efficient and accurate transcription of historical text by combining machine learning and manual review. We have open-sourced our custom annotation tool and made the reviewed images open access.

Early Detection of Parkinson’s Disease Using Machine Learning

A hallmark of Parkinson's disease is the degeneration of dopaminergic neurons in the midbrain's substantia nigra pars compacta. However, machine learning is needed to design and implement early Parkinson's disease detection. by using machine learning methods such as CNN and SVM, which can reliably identify voice signals and spiral images to identify early indications of Parkinson's disease. Using machine learning on handwriting, tremor, and gait datasets, the method addresses the shortcomings of individual analyses for a more complete diagnosis solution by investigating relationships between symptoms. This increases accuracy. When voice and spiral drawing data were combined, Parkinson's disease diagnosis accuracy showed promise, with the machine learning model successfully differentiating between unaffected patients and those who were affected. This observation suggests a viable path for precise Parkinson's disease diagnosis: an integrated method that combines machine learning techniques with data from spiral drawings and voice.

Leveraging Domain Expertise in Machine Learning for Critical Metal Prospecting in the Oslo Rift: A Case Study for Fe-Ti-P-Rare Earth Element Mineralization

Global demand for critical raw materials, including phosphorus (P) and rare earth elements (REEs), is on the rise. The south part of Norway, with a particular focus on the Southern Oslo Rift region, is a promising reservoir of Fe-Ti-P-REE resources associated with magmatic systems. Confronting challenges in mineral exploration within these systems, notably the absence of alteration haloes and distal footprints, we have explored alternative methodologies. In this study, we combine machine learning with geological expertise, aiming to identify prospective areas for critical metal prospecting. Our workflow involves processing over 400 rock samples to create training datasets for mineralization and non-mineralization, employing an intuitive sampling strategy to overcome an imbalanced sample ratio. Additionally, we convert airborne magnetic, radiometric, and topographic maps into machine learning-friendly features, with a keen focus on incorporating domain knowledge into these data preparations. Within a binary classification framework, we evaluate two commonly used classifiers: a random forest (RF) and support vector machine (SVM). Our analysis shows that the RF model outperforms the SVM model. The RF model generates a predictive map, identifying approximately 0.3% of the study area as promising for mineralization. These findings align with legacy data and field visits, supporting the map’s potential to guide future surveys.

Failure prediction and optimization for composite pressure vessel combining FEM simulation and machine learning approach

Failure assessment is one of the fundamental tasks for optimization of composite pressure vessels (CPVs). However, the extensive design space of composites usually leads to costly and repetitive work of failure assessment that hinders the design and optimization of CPVs. In this work, a combined FEM simulation and machine learning approach is used to predict the failure factors R of CPVs for further optimization. CPVs with various design parameters are automatically generated and analyzed in ABAQUS to obtain failure factors under various external loadings. Then, an assembled deep neural network (DNN) namely Multi-DNN is trained to map relations between design parameters and R of CPVs. Further, transfer learning (TL) is introduced to improve the extensibility of our approach by fine-tuning the pre-trained Multi-DNN using scarce data of new design space. Such a TL-Multi-DNN model can precisely predict R of CPVs with new design parameters. Additionally, combining the trained models and genetic algorithm, the optimization of CPVs is carried out with high efficiency and low computational cost. As a result, R of CPV with various given and new design parameters can be estimated directly by the trained models and the optimal layer sequence of CPV can be obtained efficiently.

Machine learning evaluation for identification of M-proteins in human serum.

Serum electrophoresis (SPEP) is a method used to analyze the distribution of the most important proteins in the blood. The major clinical question is the presence of monoclonal fraction(s) of antibodies (M-protein/paraprotein), which is essential for the diagnosis and follow-up of hematological diseases, such as multiple myeloma. Recent studies have shown that machine learning can be used to assess protein electrophoresis by, for example, examining protein glycan patterns to follow up tumor surgery. In this study we compared 26 different decision tree algorithms to identify the presence of M-proteins in human serum by using numerical data from serum protein capillary electrophoresis. For the automated detection and clustering of data, we used an anonymized data set consisting of 67,073 samples. We found five methods with superior ability to detect M-proteins: Extra Trees (ET), Random Forest (RF), Histogram Grading Boosting Regressor (HGBR), Light Gradient Boosting Method (LGBM), and Extreme Gradient Boosting (XGB). Additionally, we implemented a game theoretic approach to disclose which features in the data set that were indicative of the resulting M-protein diagnosis. The results verified the gamma globulin fraction and part of the beta globulin fraction as the most important features of the electrophoresis analysis, thereby further strengthening the reliability of our approach. Finally, we tested the algorithms for classifying the M-protein isotypes, where ET and XGB showed the best performance out of the five algorithms tested. Our results show that serum capillary electrophoresis combined with decision tree algorithms have great potential in the application of rapid and accurate identification of M-proteins. Moreover, these methods would be applicable for a variety of blood analyses, such as hemoglobinopathies, indicating a wide-range diagnostic use. However, for M-protein isotype classification, combining machine learning solutions for numerical data from capillary electrophoresis with gel electrophoresis image data would be most advantageous.

Assessing the quality ecology of endemic tree species in China based on machine learning models and UPLC methods: The example of Eucommiaulmoides Oliv.

Climate change significantly affects the suitability distributions of natural populations, especially relict species such as Eucommia ulmoides Oliv. (E. ulmoides). However, the driving eco-factors of the distribution of E. ulmoides and the relationship between geo-distribution patterns and quality are still unclear. Here, a new approach was established by combining machine learning models (MaxEnt, Random Forest, and Biomod2 model), ultra-performance liquid chromatography (UPLC) and spatial analysis to address this question. The results reveal that temperature is the most critical variable influencing the distribution of E. ulmoides including Bio1, Bio10, and Bio11, followed by precipitation. Multi-model predictions of suitable habitats for E. ulmoides cover an area ranging from 113.33 × 104 km2 to 285.98 × 104 km2, mainly distributed in the mountainous areas of southwestern China, including the eastern regions of the Dalou, Wuling, and Qinling-Daba Mountains. Under different future scenarios, suitable habitat shifted northward by 184.90–424.88 km, and the contraction area is greater than the expansion area, with the difference ranging from 49.28 × 104 km2 to 98.39 × 104 km2. Eight common components (e.g., chlorogenic acid and pinoresinol doglucoside) and two endemic components (caffeic acid and geniposidic acid) were found in the leaves of the northern and southern medicinal districts (SDM and NMD). The contents of eight key secondary metabolites of E. ulmoides leaves were better distributed in SDM than in NDM. Among these, Xinyang in Henan, Lu'an and Fuyang in Anhui, and the southeastern region of Zhejiang are suitable growing areas for the high-quality cultivation of E. ulmoides, which deserve special attention. The seasonal precipitation (Bio15) was negatively correlated with chlorogenic acid and pinoresinol diglucoside while longitude significantly positively correlated with pinoresinol diglucoside. Overall, this study provides a novel assessment pattern for quality assessment and effective protection of E. ulmoides.

Uncovering predictors of low hippocampal volume: Evidence from a large-scale machine-learning-based study in the UK Biobank.

Hippocampal atrophy is an established biomarker for conversion from the normal ageing process to developing cognitive impairment and dementia. This study used a novel hypothesis-free machine-learning approach, to uncover potential risk factors of lower hippocampal volume using information from the world's largest brain imaging study. A combination of machine learning and conventional statistical methods were used to identify predictors of low hippocampal volume. We run gradient boosting decision tree modelling including 2891 input features measured before magnetic resonance imaging assessments (median 9.2 years, range 4.2-13.8 years) using data from 42,152 dementia-free UK Biobank participants. Logistic regression analyses were run on 87 factors identified as important for prediction based on Shapley values. False discovery rate adjusted P-value <0.05 was used to declare statistical significance. Older age, male sex, greater height, and whole-body fat free mass were the main predictors of low hippocampal volume with the model also identifying associations with lung function and lifestyle factors including smoking, physical activity, and coffee intake (corrected P<0.05 for all). Red blood cell count and several red blood cell indices such as haemoglobin concentration, mean corpuscular haemoglobin, mean corpuscular volume, mean reticulocyte volume, mean sphered cell volume, and red blood cell distribution width were among many biomarkers associated with low hippocampal volume. Lifestyles, physical measures, and biomarkers may affect hippocampal volume, with many of the characteristics potentially reflecting oxygen supply to the brain. Further studies are required to establish causality and clinical relevance of these findings.

Anaerobic Digestion of a Curious VFA Complex Feed for Biomethane Production; A Study on ANN Modeling Optimized with Genetic Algorithm

Anaerobic digestion is a complex biological process widely used for organic waste treatment and biogas production. Understanding the intermediate stages and biochemicals is essential for effective process management. This study uses ANN modeling as well as genetic algorithm optimization to explore and predict how these intermediates behave. By scrutinizing the interactions between VFAs and CH4 production, within the context of our VFA Complex Feed characterized by unique concentrations, this model underscores the paramount significance of three VFAs: acetate, propionate, and butyrate. Notably, in this distinctive study, contrary to prior research, acetate manifests a deleterious influence on biogas production (CI = -1.92), whereas propionate (CI = +1.22) and butyrate (CI = +1.14) exhibit a favorable impact. Notably, acetate exerts the most substantial absolute influence (AAS = +4.7) when juxtaposed with other VFAs. These results support prior research, supporting its validity. By combining machine learning with theoretical knowledge, our study advances our comprehension of anaerobic digestion intermediates and offers valuable insights for optimizing the process.

Optimizing microwave-assisted synthesis of akermanite nanoparticles using citric acid as a chelating agent: A combined machine learning and experimental approach

This study aims to synthesize akermanite (Ca2MgSi2O7) using an eco-friendly and fast microwave-assisted method and understand the effect of using citric acid (CA) as a chelating agent to optimize the calcination temperature for synthesizing pure akermanite nanoparticles (along with a trace amount of merwinite) at low calcination temperatures. This study developed a new predictive model by machine learning-assisted modeling to advance the understanding of relationship between synthesis parameters and the resulting purity and particle size of akermanite. To evaluate the reliability of the developed model in predicting the purity and characteristics of the resultant Ak powder, the crystallization behavior and phase formation sequence of citric acid-containing samples and those without citric acid were experimentally investigated by differential thermal analysis (DTA) and X-ray diffraction (XRD) techniques. Following validation and training, the developed model indicates promising effects of citric acid on reducing crystallization temperature, enhancing purity, and decreasing particle size. The synthesized powders were further characterized by field emission scanning electron microscopy (FESEM), energy-dispersive X-ray spectroscopy (EDS), and transmission electron microscopy (TEM) analyses. For the first time, our results addressed the microwave-assisted synthesis as an effective method via experimental evidence and machine learning-assisted modeling that allows for the production of akermanite containing a trace amount of merwinite with the aid of citric acid at a relatively low calcination temperature of 950 °C. In vitro bioactivity evaluation supported the potential suitability of Ak powders for bone tissue engineering applications.

Predict student learning styles and suitable assessment methods using click stream

Adaptive learning, which aims to give each learner engaging, effective learning experiences, is one method of offering modified education. Adaptive learning seeks to consider the student's unique characteristics by personalizing the learning course materials and evaluation procedures. To determine the student's preferred learning strategies, we first ascertain their attributes utilizing VAK learning styles. In this study, we developed an integrated model to classify learners based on their learning activity clicks by combining machine learning algorithms like K-Nearest Neighbor (KNN), random forest (RF), and support vector machine (SVM) and Logistic regression (LR) with semantic association, which is used to help us map learning activity with VAK learning style. This enables us to classify learners, determine their preferred methods of learning, and offer the most suitable; as a result, we were able to group pupils according to their learning styles and provide the best evaluation technique or strategies. To assess the effectiveness of the suggested model, several tests were executed on the actual dataset (Open University Learning Analytics Dataset, or OULAD). According to studies, using a Random Forest algorithm, the suggested model can predict which evaluation strategy or strategies will be most effective for each student and can classify individuals with the highest degree of accuracy—98%.

One-step abductive multi-target learning with diverse noisy samples and its application to tumour segmentation for breast cancer

Recent studies have demonstrated the effectiveness of the combination of machine learning and logical reasoning, including data-driven logical reasoning, knowledge driven machine learning and abductive learning, in inventing advanced technologies for different artificial intelligence applications. One-step abductive multi-target learning (OSAMTL), an approach inspired by abductive learning, via simply combining machine learning and logical reasoning in a one-step balanced multi-target learning way, has as well shown its effectiveness in handling complex noisy labels of a single noisy sample in medical histopathology whole slide image analysis (MHWSIA). However, OSAMTL is not suitable for the situation where diverse noisy samples (DiNS) are provided for a learning task. In this paper, giving definition of DiNS, we propose one-step abductive multi-target learning with DiNS (OSAMTL-DiNS) to expand the original OSAMTL to handle complex noisy labels of DiNS. Applying OSAMTL-DiNS to tumour segmentation for breast cancer in MHWSIA, we show that OSAMTL-DiNS is able to enable various state-of-the-art approaches for learning from noisy labels to achieve more rational predictions. We released a model pre-trained with OSAMTL-DiNS for tumour segmentation in HE-stained pre-treatment biopsy images in breast cancer, which has been successfully applied as a pre-processing tool to extract tumour-associated stroma compartment for predicting the pathological complete response to neoadjuvant chemotherapy in breast cancer.