Real-world Data Mining Research Articles

A Review on Artificial Intelligence in Pharmaceutical Science

Abstract: In recent years, the use of artificial intelligence (AI) in health care has risen steadily, including a wide range of applications in the field of pharmacology. AI is now used throughout the entire continuum of pharmacology research and clinical practice and from early drug discovery to real-world data mining. The types of AI models used range from unsupervised clustering of drugs or patients aimed at identifying potential drug compounds or suitable patient populations, to supervised machine learning approaches to improve therapeutic drug monitoring. Additionally, natural language processing is increasingly used to mine electronic health records to obtain real-world data. In this mini-review, we discuss the basics of AI followed by an outline of its application in pharmacology research and clinical practice. Artificial intelligence is the upcoming technology in advance health care system. Current digitalization of medicine and availability of electronic health records (EHRs) has inspired clinical researchers and healthcare personnel to acquire artificial intelligence (AI) methodologies for big data analytics and for very large scale medical databases. The major advantage of AI is that it reduces the time that is needed for drug development and, in turn, it reduces the costs that are associated with drug development, enhances the returns on investment and may even cause a decrease in cost for the end user. A large number of researches are being carried out to improve the current available AI technology to make the pharmacy profession more efficient. The present article briefly describes the importance of AI in the process of drug development and then looks at the various AI tools that are available at the disposal of a modern-day pharmacist to aid in a more efficient functioning.

Age- and sex-specific 99th percentile upper reference limits for high-sensitivity cardiac troponin T in Chinese older people: Real-world data mining

BackgroundThis study aims to investigate the impact of age and sex on high-sensitivity cardiac troponin T (hs-cTnT) and establish 99th percentile upper reference limits (URLs) in older individuals utilizing large-scale real-world data. Methods40,530 outpatient hs-cTnT results were obtained from the laboratory database from January 1, 2018, to December 31, 2023. Our study included 4,199 elderly outpatients (aged ≥ 60) without cardiovascular disease or other heart-related chronic conditions. Nested analysis of variance was used to explore the necessity of partitioning reference intervals (RIs) by sex and age groups. RIs were established by the refineR algorithm and assessed based on ≤ 10% test results of validation data set outside the new RIs. ResultsRIs for hs-cTnT in the older population needed to be partitioned by sex and age groups ([standard deviation ratio] SDRage = 0.75; SDRsex = 0.49). URLs in older Chinese adults were 21.8 ng/L for males, 16.5 ng/L for females, and 20.7 ng/L for the overall participant group. URLs for males aged 60–69, 70–79, and ≥ 80 were 13.7, 19.4, and 31.0 ng/L, respectively. Female values were 10.1, 17.2, and 22.0 ng/L. Importantly, manufacturer-reported RIs do not suffice for Chinese individuals aged ≥ 70. Validation data showed that 2.7–5.2% of test results fell outside the new RIs, confirming the validity of the results. ConclusionThis study establishes age- and sex-specific 99th percentile URLs for hs-cTnT in Chinese older individuals, thereby enhancing the accuracy of clinical assessments.

The safety signal detection and analysis of monoclonal antibodies against SARS-CoV-2 based on real-world evidence - the suitable selectivity for different populations.

Bebtelovimab (BEB), Tixagevimab/Cilgavimab (TIX/CIL), and Sotrovimab (SOT) are important agents against the severe acute respiratory syndrome coronavirus 2-Omicron strain. However, due to their short duration of application, little is known about their safety profiles. This research aimed to explore the safety profile of these monoclonal antibodies (mAbs) via real-world evidence databases and data mining tools. Safety reports were retrieved from the database of the U.S. Food and Drug Administration (FDA) Adverse Event Reporting System from April 2022 to March 2023. To detect the safety signal, the disproportionality analysis was performed using the reporting odds ratio method. SOT had the greatest proportion of "skin and subcutaneous tissue disorders" and "disorders of investigations"; BEB showed significant associations with "gastrointestinal disorders" and "nervous system disorders"; TIX/CIL had the weakest correlation with "skin and subcutaneous tissue disorders" and "general disorders and administration site conditions". Furthermore, there were still other signals related to nervous system disorders, gastrointestinal disorders only caused by BEB. TIX/CIL has been reported solely to be associated with multiple types of cardiovascular disorders. As for SOT alone, signals were strongly related to infusion reactions and hypersensitivity. In summary, SOT may be unsuitable for allergic patients and may lead to abnormal test results. BEB showed the highest correlations with gastrointestinal and neuropsychiatric events. In addition, its infusion reactions should also be noted. TIX/CIL can lead to a variety of cardiovascular events.

Abstract 2120: Exploring the genomic landscape of colorectal cancer for enabling clinical genomics informed and real-world data based clinical development

Abstract Background: Activation of the MAPK-signaling by RAS- (KRAS or NRAS) alterations is an important driver event in colorectal cancer (CRC) tumorigenesis that determines treatment selection (1).Real-world data (RWD) mining is an important tool to inform patient selection and stratification (2,3). Here, RWD genomics data mining was conducted using the Foundation Medicine database to explore RAS and BRAF-alterations in CRC patients and to investigate the general alterations’ landscape - including rare alterations. The prevalence of microsatellite instability (MSI) and tumor mutational burden (TMB) were also explored. Methods: The prevalence of gene alterations and their co-occurrence with RAS/BRAF-alterations in tumor tissue of CRC patients (N=42800) were investigated using the FoundationInsightsTM web platform, which includes harmonized results from the FoundationOneCDx, FoundationOneHeme, and PD-L1 assays from 2012 to March 2022. Data was collected from FMI solid and heme tests from 2012 to March 2022. Alterations that render a particular gene ‘altered’ included single nucleotide variants (SNVs) or insertion-deletion mutations (Indels), collectively referred to as short variants (SVs), rearrangements (RE) and copy number events (CN). TMB (cut-off 10 mut/MB) and MSI status were also characterized. Results: Altogether, 36.9% of CRC samples were RAS/BRAF-WT and 63.1% had a known or likely functional RAS/BRAF-alteration, respectively. Of all KRAS and NRAS alterations (SV, CN and RE regardless of annotation as functional or presence in a hotspot), SVs in exons 2, 3 and 4 constituted 96.1% and 94.2% of all alterations, respectively. Relative to all samples the potentially targetable KRAS p.G12C, p.G12D and p.G12V mutations were present in 3.6%, 15.1% and 10.2% of samples, respectively. The p.G12E variant was present in less than ten samples. The five most frequently altered potentially targetable genes were PIK3CA (19.6%), PTEN (8.7%), ERBB2 (4.7%), EGFR (2.5%) and FGFR1 (2.0%). NTRK-alterations (NTRK1/2/3) were present in 0.9%, ROS1-alterations in 0.4%, RET-alterations in 0.6% and ALK-alterations in 0.5% of samples, respectively. The prevalence of immuno-oncology biomarkers was also explored; overall, 9.6% of samples were TMB-high, 6.1% MSI-high and 0.6% POLE-altered. MSI-high or TMB-high samples were significantly more abundant in the RAS/BRAF-altered subset. In the RAS/BRAF-WT subset 4.6% were MSI-high and 7.9% TMB-high, whereas in the RAS/BRAF-altered subset 7.1% were MSI-high and 10.6% TMB-high. Conclusion: RWD based clinical genomic profiling confirms that the vast majority of RAS-alterations in CRC are SVs in exons 2, 3 and 4. Both TMB-high and MSI-high samples are significantly more abundant in the RAS/BRAF-altered cohort. RWD confirms that PI3K-, ERBB- and FGFR-signaling pathways are frequently altered in CRC. RWD based clinical genomic profiling has the potential to support comprehensive clinical study design as well as patient selection/stratification. Citation Format: Markus Schulze, XiaoZhe (Janet) Wang, Jawad Hamad, Julia F. Hopkins, Julia F. Hopkins, Jürgen Scheuenpflug, Zheng Feng. Exploring the genomic landscape of colorectal cancer for enabling clinical genomics informed and real-world data based clinical development [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2120.

Forecasting Quality of Service (QoS) Parameters of IoT based Web Services with an Artificial Electric Field Algorithm trained Artificial Neural Network

Development of an accurate forecasting model for effective prediction of Quality of Service (QoS) parameters of inter-net of things (IoT) based web services is highly desired, such that it improves service management and user experience. Mostly, QoS parameters are volatile in nature which make the IoT based service and recommendation process chal-lenging. Artificial neural network (ANN) based models are found to be worthy in modeling and forecasting nonlinear QoS parameter sequences. However, improper tuning of ANN parameters with conventional training algorithms may lead to a suboptimal model. Nature-inspired optimization methods are found suitable in fine tuning ANN parameters and have shown proficient results on real-world data mining problems. There is lack of such models for QoS parameters prediction that need to be explored. We develop an Artificial Electric Field Algorithm (AEFA) trained ANN (AEFANN) model for effective and accurate prediction of QoS parameters where AEFA is used to search an optimal ANN structure. The optimal ANN structure is achieved by AEFA through an evolutionary process. Two real-world IoT enabled web service datasets are used for evaluating effectiveness of AEFANN in terms of three performance metrics. Experimental procedures and comparative studies are conducted to establish the superiority of the proposed approach over four other similar forecasts. AEFANN obtained relative worth values of 4.13% ~ 69.12 % (5-min granularity) and 43.32% ~ 80.3 % (1-hr granularity) from SERVICE 1 dataset. Similarly, it obtained relative worth values of 7.25% ~ 65.57 % (5-min granularity) and 43.38% ~ 72.43 % (1-hr granularity) from SERVICE 2 dataset when compared to oth-er models. This is a significant improvement over comparative existing similar model.

Trends in tolvaptan prescription and the association between hypernatremia and aging in tolvaptan-treated patients in Japan: Real-world data mining using Japanese databases.

To identify the trends in tolvaptan prescription and the association between aging and tolvaptan-induced hypernatremia. A health insurance claims database and a spontaneous adverse drug reaction database were used. Of all patients who had been prescribed tolvaptan, the proportion of patients aged 60 - 79 years and ≥ 80 years was consistent at ~ 40%. Moreover, the prescription frequency of tolvaptan increased over time for patients in the same age groups. The adjusted reporting odds ratio of tolvaptan-induced hypernatremia was 5.54 (95% confidence interval, 3.31 - 9.25) in patients aged ≥ 60 years from among all patients and 2.09 (95% confidence interval, 1.59 - 2.75) in those aged ≥ 80 years from among those aged ≥ 60 years. It may be necessary to be aware of hypernatremia in elderly patients who are expected to have increased prescriptions of tolvaptan.

Real-world data mining meets clinical practice: Research challenges and perspective.

DESY-22-153.

An innovative approach based on real-world big data mining for calculating the sample size of the reference interval established using transformed parametric and non-parametric methods

BackgroundCurrently, the direct method is the main approach for establishment of reference interval (RI). However, only a handful of studies have described the effects of sample size on establishment of RI and estimation of sample size. We describe a novel approach for estimation of the sample size when establishing RIs using the transformed parametric and non-parametric methods.MethodsA total of 3,697 healthy participants were enrolled in this study. We adopted a two-layer nested loop sample size estimation method to determine the effects of sample size on RI, using thyroid-related hormone as an example. The sample size was selected as the calculation result when the width of the confidence interval (CI) of the upper and lower limit of the RI were both stably < 0.2 times the width of RI. Then, we calculated the sample size for establishing RIs via transformed parametric and non-parametric methods for thyroid-related hormones.ResultsSample sizes for thyroid stimulating hormone (TSH), as required by parametric and non-parametric methods to establish RIs were 239 and 850, respectively. Sample sizes required by the transformed parametric method for free triiodothyronine (FT3), free thyroxine (FT4), total triiodothyronine (TT3) and total thyroxine (TT4) were all less than 120, while those required by the non-parametric method were more than 120.ConclusionWe describe a novel approach for estimating sample sizes for establishment of RI. A corresponding open-source code has been developed and is available for applications. The established method is suitable for most analytes, with evidence based on thyroid-related hormones indicating that different sample sizes are required to establish RIs using different methods for analytes with different variations.

TRANSLATIONAL RESEARCH ETHICS

The purpose of translational research is to test novel therapeutic strategies and drugs throughout basic and clinical research. The basic concept of translational research is ‘Bench to Bedside’, defining the collaboration between clinicians and basic scientists. That means translational research involves synergistic partnerships between industry, foundations, and public health services. The collaboration between partners provides a network structure that uses multiple methods and technologies to harmonize parameters such as complexity, expertise, safety, and efficiency with national regulations, institutional policies, and best practices that govern cell, animal, and human research. One might think that the ethics of translational research is simple because there are extensive rules regarding basic and clinical research ethics. Nevertheless, translational research brings new tools to the scene such as employing biomarkers based on omics technology and genomics. On the other hand, data science using real-world data, big data, data mining, and artificial intelligence is another huge leverage for translational research bringing new ethical and legal issues, especially concerning data protection. Therefore, new ethical issues arise that have not been addressed before. As the name reflects, translational research ethics should be considered as the means of a literal translation as well. Ethics should serve to translate the language used by different disciplines and moreover by all stakeholders including social areas from the public to the governmental bodies etc. to translate these different languages and attitudes to each other and make the science more transparent and reliable.

M267 Establishment of biological variation and age-related reference interval for thyroid hormones in the elderly using real-world big data mining

M267 Establishment of biological variation and age-related reference interval for thyroid hormones in the elderly using real-world big data mining

A linear programming approach to multiple instance learning

Multiple instance learning (MIL) aims to classify objects with complex structures and covers a wide range of real-world data mining applications. In MIL, objects are represented by a bag of instances instead of a single instance, and class labels are provided only for the bags. Some of the earlier MIL methods focus on solving MIL problem under the standard MIL assumption, which requires at least one positive instance in positive bags and all remaining instances are negative. This study proposes a linear programming framework to learn instance level contributions to bag label without emposing the standart assumption. Each instance of a bag is mapped to a pseudo-class membership estimate and these estimates are aggregated to obtain the bag-level class membership in an optimization framework. A simple linear mapping enables handling various MIL assumptions with adjusting instance contributions. Our experiments with instance-dissimilarity based data representations verify the effectiveness of the proposed MIL framework. Proposed mathematical models can be solved efficiently in polynomial time.

Automatic recommendation of feature selection algorithms based on dataset characteristics

Feature selection in real-world data mining problems is essential to make the learning task efficient and more accurate. Identifying the best feature selection algorithm, among the many available, is a complex activity that still relies heavily on human experts or some random trial-and-error procedure. Thus, the automated machine learning community has taken some steps towards the automation of this process. In this paper, we address the metalearning challenge of recommending feature selection algorithms by proposing a novel meta-feature engineering model. Our model considers a broad collection of meta-features that enable the study of the relationship between the dataset properties and the feature selection algorithm performance in terms of several criteria. We arrange the input meta-features into eight categories: (i) simple, (ii) statistical, (iii) information-theoretical, (iv) complexity, (v) landmarking, (vi) based on symbolic models, (vii) based on images, and (viii) based on complex networks (graphs). The target meta-features emerge from a multi-criteria performance measure, based on five individual performance indexes, that assesses feature selection methods grounded in information, distance, dependence, consistency, and precision measures. We evaluate our proposal using a recently developed framework that extracts the input meta-features from 213 benchmark datasets, and ranks the assessed feature selection algorithms, to fill in the target meta-features in meta-bases. This evaluation uses five state-of-the-art classification methods to induce recommendation models from meta-bases: C4.5, Random Forest, XGBoost, ANN, and SVM. The results showed that it is possible to reach an average accuracy of up to 90% applying our meta-feature engineering model. This work is the first to use an extensive empirical evaluation to provide a careful discussion of the strengths and limitations of more than 160 meta-features. These meta-features, while designed to aid the task of feature selection algorithm recommendation, can readily be employed in other metalearning scenarios. Therefore, we believe our findings are a valuable contribution to the fields of automated machine learning and data mining, as well as to the feature extraction and pattern recognition communities.

Establishment of variation source and age-related reference interval models for 22 common biochemical analytes in older people using real-world big data mining.

the ageing population has increased in many countries, including China. However, reference intervals (RIs) for older people are rarely established because of difficulties in selecting reference individuals. Here, we aimed to analyse the factors affecting biochemical analytes and establish RI and age-related RI models for biochemical analytes through mining real-world big data. data for 97,220 individuals downloaded from electronic health records were included. Three derived databases were established. The first database included 97,220 individuals and was used to build age-related RI models after identifying outliers by the Tukey method. The second database consisted of older people and was used to establish variation source models and RIs for biochemical analytes. Differences between older and younger people were compared using the third database. sex was the main source of variation of biochemical analytes for older people in the variation source models. The distributions of creatinine and uric acid were significantly different in the RIs of biochemical analytes for older people established according to sex. Age-related RI models for biochemical analytes that were most affected by age were built and visualized, revealing various patterns of changes from the younger to older people. the study analysed the factors affecting biochemical analytes in older people. Moreover, RI and age-related RI models of biochemical analytes for older people were established to provide important insight into biological processes and to assist clinical use of various biochemical analytes to monitor the status of various diseases for older people.

Navo Minority Over-sampling Technique (NMOTe): A Consistent Performance Booster on Imbalanced Datasets

Imbalanced data refers to a problem in machine learning where there exists unequal distribution of instances for each classes. Performing a classification task on such data can often turn bias in favour of the majority class. The bias gets multiplied in cases of high dimensional data. To settle this problem, there exists many real-world data mining techniques like over-sampling and under-sampling, which can reduce the Data Imbalance. Synthetic Minority Oversampling Technique (SMOTe) provided one such state-of-the-art and popular solution to tackle class imbalancing, even on high-dimensional data platform. In this work, a novel and consistent oversampling algorithm has been proposed that can further enhance the performance of classification, especially on binary imbalanced datasets. It has been named as NMOTe (Navo Minority Oversampling Technique), an upgraded and superior alternative to the existing techniques. A critical analysis and comprehensive overview on the literature has been done to get a deeper insight into the problem statements and nurturing the need to obtain the most optimal solution. The performance of NMOTe on some standard datasets has been established in this work to get a statistical understanding on why it has edged the existing state-of-the-art to become the most robust technique for solving the two-class data imbalance problem.

DCE -miner: an association rule mining algorithm for multimedia based on the MapReduce framework

The amount of multimedia data has grown rapidly because of improvements in data collection and storage technologies. The association rule mining (ARM) technique is a type of data mining method widely used to extract useful information from data warehouses. In real-world big data applications, fast and effective data mining algorithms are emerging as a valuable approach. In this paper, we propose DCE-Miner, a fast association rule mining algorithm with low memory requirements based on the MapReduce framework. In the precomputation phase, we split large datasets into equal-sized smaller ones using data division method. In the frequent K-itemsets mining phase, the mappers read the small datasets and distribute the data to reducers based on the closed set characteristics associated with each partition. The reducers use bitmaps to accelerate the computation speed and store the possible frequent 2-itemsets to reduce future computation. Extensive experimental results show that on large-scale datasets with up to 40 million transactions, DCE-Miner achieves better performance and is more robust with respect to dataset sizes and support level than are the current algorithms.

Granularity and Entropy of Intuitionistic Fuzzy Information and Their Applications.

A granular structure of intuitionistic fuzzy (IF) information presents simultaneously the similarity and diversity of samples. However, this structural representation has rarely displayed its technical capability in data mining and information processing due to the lack of suitable constructive methods and semantic interpretation for IF information with regard to real data. To pursue better performance of the IF-based technique in real-world data mining, in this article, we examine information granularity, information entropy of IF granular structures, and their applications to data reduction of IF information systems. First, several types of partial-order relations at different hierarchical levels are defined to reveal the granularity of IF granular structures. Second, the granularity invariance between different IF granular structures is characterized by using relational mappings. Third, Shannon's entropies are generalized to IF entropies and their relationships with the partial-order relations are addressed. Based on the theoretical analysis above, the significance of intuitionistic attributes using the information measures is then introduced and the information-preserving algorithm for data reduction of IF information systems is constructed. Finally, by inducing substantial IF relations from public datasets that take both the similarity/diversity between the samples from the same/different classes into account, a collection of numerical experiments is conducted to confirm the performance of the proposed technique.

Association between oral anticoagulants and osteoporosis: Real-world data mining using a multi-methodological approach.

Introduction: Warfarin and direct oral anticoagulants (DOACs) have been widely used in antithrombotic therapy. Although warfarin use has been suspected to be associated with osteoporosis risk, several studies have shown otherwise. Conversely, a few reports have found an association between DOACs and osteoporosis. This study therefore clarifies the association between oral anticoagulants and osteoporosis by analyzing real-world data using different methodologies, algorithms, and databases.Methods: Real-world data from the US Food and Drug Administration Adverse Event Reporting System (FAERS; 2004-2016) and Japanese administrative claims database (2005-2017; JMDC Inc., Tokyo) were used. Reporting odds ratio (ROR) and information component (IC) were calculated through disproportionality analysis (DPA) using reports recorded in the FAERS. Sequence symmetry analysis (SSA) was employed to calculate the adjusted sequence ratio (SR) using the JMDC Claims Database. For the adjusted SR and ROR, a significant signal was detected when the lower limit of the two-sided 95% confidence interval (CI) was more than 1. For the IC, a significant signal was detected when the lower limit of the 95% CI was more than 0.Results: DPA for warfarin found significant signals for osteoporosis in ROR (1.43, 95% CI: 1.32-1.54) and IC (0.50, 95% CI: 0.39-0.61). SSA showed a significant association between warfarin use and osteoporosis or bisphosphonate use. Moreover, a significant association was observed in males and females, albeit only for warfarin.Conclusion: Multi-methodological data mining revealed that warfarin use, not DOACs, is significantly associated with osteoporosis regardless of sex difference.

A novel approach to ensemble learning in distributed data mining

Several data mining techniques have been proposed to take out hidden information from databases. Data mining and knowledge extraction becomes challenging when data is massive, distributed and heterogeneous. Classification is an extensively applied task in data mining for prediction. Huge numbers of machine learning techniques have been developed for the purpose. Ensemble learning merges multiple base classifiers to improve the performance of individual classification algorithms. In particular, ensemble learning plays a significant role in distributed data mining. So, study of ensemble learning is crucial in order to apply it in real-world data mining problems. We propose a technique to construct ensemble of classifiers and study its performance using popular learning techniques on a range of publicly available datasets from biomedical domain.

A Taxonomy of Data Mining Problems

Much of the research in data mining and knowledge discovery has focused on the development of efficient data mining algorithms. Researchers and practitioners have developed data mining techniques to solve diverse real-world data mining problems. But there is no single source that identifies which techniques solve what problems and how, the advantages and limitations, and real-life use-cases. Lately, identifying data mining techniques and corresponding problems that they solve has drawn significant attention. In this paper, the author describes the progress made in developing data mining techniques and then classify them in terms of data mining problems taxonomy to help assist practitioners in using appropriate data mining techniques that solve business problems. This will allow researchers to expand the body of knowledge in this discipline. This article proposes a data mining problems taxonomy based on data mining techniques being used. Prominent data mining problems include classification, optimization, prediction, partitioning, relationship, pattern matching, recommendation, ranking, sequential patterns and anomaly detection. The data mining techniques that are used to solve these data mining problems in general fall under top 10 data mining algorithms.

A pure array structure and parallel strategy for high-utility sequential pattern mining

High-utility sequential pattern mining (HUSPM) is the task of discovering all sequential patterns in a sequence database whose utility values are equal to or greater than a given minimum utility threshold. HUSPM has become increasingly important in many real-world data mining applications, such as market basket data analysis, weblog mining, and bio-medical gene data analysis, which considers co-occurrence values and quantity, utility (e.g., profit or cost) and time. Current approaches in the literature for HUSPM use the utility matrix to store a sequence database in the memory. Unfortunately, the utility matrix consumes a large amount of main memory. To address this issue, we introduce a pure array structure that reduces the memory consumption when compared to the utility matrix. In addition, HUSPM is also challenged with the downward closure property (DCP) to prune the search space. Recently, HUSPM algorithms have used the upper bound of utility values as the DCP. However, it is usually higher than the actual utility of patterns. Thus, these algorithms may generate many candidate patterns. The large search space leads to poor performance due to excessive runtime and memory usage. One of the reasons is the number of candidate patterns is proportional to the number of requisite projected database scans for calculating their actual utilities. In this paper, we present a novel pruning strategy that efficiently prunes non-HUSPs and significantly reduces the search space compared to the state-of-the-art HUS-Span algorithm. Moreover, we propose a parallel strategy to speed up the mining process. Then, we propose two algorithms which are the pure Array structure for High-utility Sequential (AHUS) pattern mining and AHUS parallel mining (AHUS-P). The AHUS-P algorithm uses shared memory to parallelize the mining process. It concurrently identifies HUSPs based on the advantages of the multi-core processor architecture. The experimental results show that AHUS and AHUS-P can efficiently and effectively discover all HUSPs. Both the proposed algorithms outperform the state-of-the-art HUS-Span algorithm in terms of runtime, memory usage, and scalability for all experimental datasets.