Rule Discovery Research Articles

Discovering Top-k Relevant and Diversified Rules

This paper studies the problem of discovering top- k relevant and diversified rules. Given a real-life dataset, it is to mine a set of k rules that are as close to users' interest as possible, and meanwhile, as diverse to each other as possible. It aims to reduce excessive irrelevant rules commonly returned by rule discovery. As a testbed, we consider Entity Enhancing Rules (REEs), which subsume popular data quality rules as special cases. We train a relevance model to learn users' prior knowledge, rank rules based on users' need, and propose four diversity measures to assess the diversity between rules. Based on these measures, we formulate a new discovery problem. We show that the bi-criteria discovery problem is NP-complete and hard to approximate. This said, we develop a practical algorithm for the problem, and prove its approximation bounds under certain conditions. Moreover, we develop optimization techniques to speed up the process, and parallelize the algorithm such that it guarantees to reduce runtime when given more processors. Using real-life data, we empirically verify that on average, the top-10 REEs discovered by our algorithm is able to catch 77.5% of errors detected by the entire set Σ all of REEs and achieve F_1 = 0.74 for real error detection; moreover, discovering top-ranked REEs is 62.4X faster than mining Σ all .

A Recommendation System for Trigger-Action Programming Rules via Graph Contrastive Learning.

Trigger-action programming (TAP) enables users to automate Internet of Things (IoT) devices by creating rules such as "IF Device1.TriggerState is triggered, THEN Device2.ActionState is executed". As the number of IoT devices grows, the combination space between the functions provided by devices expands, making manual rule creation time-consuming for end-users. Existing TAP recommendation systems enhance the efficiency of rule discovery but face two primary issues: they ignore the association of rules between users and fail to model collaborative information among users. To address these issues, this article proposes a graph contrastive learning-based recommendation system for TAP rules, named GCL4TAP. In GCL4TAP, we first devise a data partitioning method called DATA2DIV, which establishes cross-user rule relationships and is represented by a user-rule bipartite graph. Then, we design a user-user graph to model the similarities among users based on the categories and quantities of devices that they own. Finally, these graphs are converted into low-dimensional vector representations of users and rules using graph contrastive learning techniques. Extensive experiments conducted on a real-world smart home dataset demonstrate the superior performance of GCL4TAP compared to other state-of-the-art methods.

US Supreme Court holds that copyright law’s discovery rule does not impose a 3-year limit on infringement damages

US Supreme Court holds that copyright law’s discovery rule does not impose a 3-year limit on infringement damages

Rock: Cleaning Data with both ML and Logic Rules

We demonstrate Rock, a system for cleaning relational data. Rock highlights the following unique features: (1) it extends logic rules by embedding machine learning models as predicates, to benefit from both ML and logic deduction; (2) it supports entity resolution, conflict resolution, timeliness deduction and missing data imputation in a unified process; and (3) it provides parallelly scalable algorithms for rule discovery, error detection and error correction, in batch and incremental modes. We will demonstrate Rock for its (a) easy-to-use interface, (b) scalability when cleaning large datasets, (c) accuracy for detecting and correcting errors across multiple tables, and (d) applications at banks and HR departments.

KGRED: Knowledge-graph-based rule discovery for weakly supervised data labeling

In weakly supervised learning, labeling rules can automatically label data to train models. However, due to insufficient prior knowledge, rule discovery often suffers from semantic drift. Since misclassified rules are generated from wrongly matched sentences, the sentences matched by rules shift from the target labels to other labels. It is worth noting that rules do not exist in isolation. The multi-dimensional semantic associations among rules can impose semantic constraints for rule generation, as well as enrich the semantic information of rules for rule matching. Therefore, we propose a Knowledge-Graph-based RulE Discovery method (KGRED), which can leverage the multi-dimensional semantic associations among rules to alleviate semantic drift in rule discovery. Specifically, to decrease misclassified rules, we design a label-aware rule generation approach to attentively propagate prior knowledge from seed rules to candidate rules based on rule KG. To reduce wrongly-matched sentences, we present a cross-attention-based semantic matching mechanism to refine the semantic information of sentences while enriching that of rules. Moreover, we propose an inconsistency-directed active learning strategy to verify rules that perform inconsistently in rule generation and matching. Experiments on two public datasets prove that KGRED can achieve at least 5.1 % gain in F1 score compared to state-of-the-art methods.

Can there be responsible AI without AI liability? Incentivizing generative AI safety through ex-post tort liability under the EU AI liability directive

Abstract In Europe, the governance discourse surrounding artificial intelligence (AI) has been predominantly centred on the AI Act, with a proliferation of books, certification courses, and discussions emerging even before its adoption. This narrow focus has overshadowed other crucial regulatory interventions that promise to fundamentally shape AI. This article highlights the proposed EU AI liability directive (AILD), the first attempt to harmonize general tort law in response to AI-related threats, addressing critical issues such as evidence discovery and causal links. As AI risks proliferate, this article argues for the necessity of a responsive system to adequately address AI harms as they arise. AI safety and responsible AI, central themes in current regulatory discussions, must be prioritized, with ex-post liability in tort playing a crucial role in achieving these objectives. This is particularly pertinent as AI systems become more autonomous and unpredictable, rendering the ex-ante risk assessments mandated by the AI Act insufficient. The AILD’s focus on fault and its limited scope is also inadequate. The proposed easing of the burden of proof for victims of AI, through enhanced discovery rules and presumptions of causal links, is insufficient in a context where Large Language Models exhibit unpredictable behaviours and humans increasingly rely on autonomous agents for complex tasks. Moreover, the AILD’s reliance on the concept of risk, inherited from the AI Act, is misplaced, as tort liability intervenes after the risk has materialized. However, the inherent risks in AI systems could justify EU harmonization of AI torts in the direction of strict liability. Bridging the liability gap will enhance AI safety and responsibility, better protect individuals from AI harms, and ensure that tort law remains a vital regulatory tool.

Temporal Association Rule Mining: Race-Based Patterns of Treatment-Adverse Events in Breast Cancer Patients Using SEER-Medicare Dataset.

Disparities in the screening, treatment, and survival of African American (AA) patients with breast cancer extend to adverse events experienced with systemic therapy. However, data are limited and difficult to obtain. We addressed this challenge by applying temporal association rule (TAR) mining using the SEER-Medicare dataset for differences in the association of specific adverse events (AEs) and treatments (TRs) for breast cancer between AA and White women. We considered two categories of cancer care providers and settings: practitioners providing care in the outpatient units of hospitals and institutions and private practitioners providing care in their offices. We considered women enrolled in the Medicare fee-for-service option at age 65 who qualified by age and not disability, who were diagnosed with breast cancer with attributed patient factors of age and race, marital status, comorbidities, prior malignancies, prior therapy, disease factors of stage, grade, and ER/PR and Her2 status and laterality. We included 141 HCPCS drug J codes for chemotherapy, biotherapy, and hormone therapy drugs, which we consolidated into 46 mechanistic categories and generated AE data. We consolidated AEs from ICD9 codes into 18 categories associated with breast cancer therapy. We applied TAR mining to determine associations between the 46 TR and 18 AE categories in the context of the patient categories outlined. We applied the spark.mllib implementation of the FPGrowth algorithm, a parallel version called PFP. We considered differences of at least one unit of lift as significant between groups. The model's results demonstrated a high overlap between the model's identified TR-AEs associated set and the actual set. Our results demonstrate that specific TR/AE associations are highly dependent on race, stage, and venue of care administration. Our data demonstrate the usefulness of this approach in identifying differences in the associations between TRs and AEs in different populations and serve as a reference for predicting the likelihood of AEs in different patient populations treated for breast cancer. Our novel approach using unsupervised learning enables the discovery of association rules while paying special attention to temporal information, resulting in greater predictive and descriptive power as a patient's health and life status change over time.

Deciphering Patterns in a small scale Case Analysis Study of the Apriori Algorithm in Market Basket Analysis using machine learning tools

Analysis of Transaction Database Using Apriori Algorithm. This study examines the use of the Apriori algorithm for analysing transaction databases. The Apriori algorithm is a fundamental technique in data mining that allows for the efficient discovery of frequent patterns and association rules in large datasets. The Apriori algorithm employs a two-step approach. Initially, it identifies frequent items in the database based on a user-defined minimum support threshold. Subsequently, it generates association rules that describe relationships between these frequent items based on metrics such as confidence and lift. This paper provides an in-depth explanation of the Apriori algorithm, emphasizing its strengths and limitations. Additionally, it presents various applications of the Apriori algorithm in real-world scenarios, such as shopping cart analysis, cross-selling and upselling, and customer segmentation. The significance of this study lies in its comprehensive analysis of the Apriori algorithm and its practical relevance in diverse data mining tasks. It serves as a valuable resource for researchers, practitioners, and anyone seeking to understand the fundamentals and applications of association rule mining. Keywords: Datamining, OLTP, Apriori algorithm, Item sets and Market Basket Analysis

Round-up of United States copyright cases 2023

Abstract The US Supreme Court and lower court provided further guidance on fair use defense. The US Copyright Office provided guidance on AI as an author. The Court of Appeals provided guidance on substantial similiarty, server test, abstraction-filtration-comparison, and discovery rule.

Привлечение дополнительных знаний о предметной области в задачах машинного обучения

The study deals with the concideration of a range of issues related to the involving additional knowledge about subject domain when solving machine learning problems. The techniques for accounting such knowlwdge based on the modification of classical methods of classification, clustering, associative rule discovery are described. It is concluded that the analysis of bachgroung knowledge is able to increase reliability and accuracy of classical machine learning methods, although their modification, taking into account additional constraints, sometimes turns out to be a rather time-consuming procedure. In addition, different types of machine learning problems require different types of additional constraints to increase the reliability and accuracy of their results. This provides certaint difficulties in solving complex problems that require gradual involvement of various types of additional constraints.

Formal Software Architecture Rule Learning: A Comparative Investigation between Large Language Models and Inductive Techniques

This paper explores the application of inferring software architecture rules from examples using Machine Learning (ML). We investigate different methods from Inductive Rule Learning and utilize Large Language Models (LLMs). Traditional manual rule specification approaches are time-consuming and error-prone, motivating the need for automated rule discovery. Leveraging a dataset of software architecture instances and a meta-model capturing implementation facts, we used inductive learning algorithms and LLMs to extract meaningful rules. The induced rules are evaluated against a predefined hypothesis and their generalizability across different system subsets is investigated. The research highlights the capabilities and limitations of ML-based rule learning in the area of software architecture, aiming to inspire further innovation in data-driven rule discovery for more intelligent software architecture practices.

Capturing More Associations by Referencing External Graphs

This paper studies association rule discovery in a graph G 1 by referencing an external graph G 2 with overlapping information. The objective is to enrich G 1 with relevant properties and links from G 2 . As a testbed, we consider Graph Association Rules (GARs). We propose a notion of graph joins to enrich G 1 by aligning entities across G 1 and G 2 . We also introduce a graph filtering method to support graph joins, by fetching only the data of G 2 that pertains to the entities of G 1 , to reduce noise and the size of the fused data. Based on these we develop a parallel algorithm to discover GARs across G 1 and G 2 . Moreover, we provide an incremental GAR discovery algorithm in response to updates to G 1 and G 2 . We show that both algorithms guarantee to reduce parallel runtime when given more processors. Better yet, the incremental algorithm is bounded relative to the batch one. Using real-life and synthetic data, we empirically verify that the methods improve the accuracy of association analyses by 30.4% on average, and scale well with large graphs.

Scalable Action Mining Hybrid Method for Enhanced User Emotions in Education and Business Domain

Education sector ,Business field,Medical domain and SocialMedia,huge amounts of data in a single day . Mining this data can provide a lot of meaningful insights on how to improve user experience in social media, users engage in these domains collect and cherish the data as they hope to find patterns and trends and the golden nuggets that help them to accomplish their goal. For example: How to improve student learning; how to increase business profit ability; how to improve user experience in social media; and how to heal patients and assists hospital administrators. Action Rule Miningmines actionable patterns which are hidden in various datasets. Action Rules provide actionable suggestions on how to change the state of an object from an existing state to a desired state for the benefit of the user. There are two major frameworks in the literature of Action Rule mining namely Rule-Based method where the extraction of Action Rules is dependent on the pre-processing step of classification rule discovery and Object-Based method where it extracts the Action Rules directly from the database without the use of classification rules.Hybrid Action rule mining approach combines both these frame works and generates complete set of Action Rules. The hybrid approach shows significant improvement in terms computational performance over the Rule-Based and Object-Based approach. In this work we propose a novel Modified Hybrid Action rule method with Partition Threshold Rho, which further improves the computational performance with large datasets.

Implementation of a Product-Recommender System in an IoT-Based Smart Shopping Using Fuzzy Logic and Apriori Algorithm

The Internet of Things (IoT) has recently become important in accelerating various functions, from manufacturing and business to healthcare and retail. A recommender system can handle the problem of information and data buildup in IoT-based smart commerce systems. These technologies are designed to determine users' preferences and filter out irrelevant information. Identifying items and services that customers might be interested in and then convincing them to buy is one of the essential parts of effective IoT-based smart shopping systems. Due to the relevance of product-recommender systems from both the consumer and shop perspectives, this article presents a new IoT-based smart product-recommender system based on an apriori algorithm and fuzzy logic. The suggested technique employs association rules to display the interdependencies and linkages among many data objects. The most common use of association rule discovery is “shopping cart analysis.” Customers' buying habits and behavior are studied based on the numerous goods they place in their shopping carts. As a result, the association rules are generated using a fuzzy system. The apriori algorithm then selects the product based on the provided fuzzy association rules. The results revealed that the suggested technique had achieved acceptable results in terms of mean absolute error, root-mean-square error, precision, recall, diversity, novelty, and catalog coverage when compared to cutting-edge methods. Finally, the method helps increase recommender systems' diversity in IoT-based smart shopping.

Online semi-supervised learning of composite event rules by combining structure and mass-based predicate similarity

Symbolic event recognition systems detect event occurrences using first-order logic rules. Although existing online structure learning approaches ease the discovery of such rules in noisy data streams, they assume the existence of fully labelled training data. Splice is a recent online graph-based approach that estimates the labels of unlabelled data and makes it possible to learn such rules from semi-supervised training sequences of logical interpretations. However, Splice labelling depends significantly on the metric used to compute the distances of unlabelled examples to their labelled counterparts. Moreover, there is no guarantee about the quality of the labelling found in the local graphs that are built as the data stream in. In this paper, we propose a new online learning method, which includes an enhanced hybrid measure that combines an optimised structural distance, and a data-driven one. The former is guided by feature selection targeted to kNN classification, while the latter is a mass-based dissimilarity. Additionally, the enhanced Splice method, improves the graph construction process, by storing a synopsis of the past, in order to achieve more informed labelling on the local graphs. We evaluate our approach by learning Event Calculus theories for the tasks of human activity recognition, maritime monitoring, and fleet management. The evaluation suggests that our approach outperforms its predecessor, in terms of inferring the missing labels and improving the predictive accuracy of the underlying structure learning system.

Machine-Learning Based Prediction of Lung Cancer

The incidence of lung cancer has now exceeded all other types of cancer globally, making it the leading cause of cancer-related deaths. Compared to other types of cancer, lung cancer has a poor prognosis and high mortality rate, making it challenging for humans to accurately predict its incidence rates. The goal of the study is to implement machine learning techniques for the early detection of lung cancer, which can increase patient survival rates. To find the most important factors and predict the possibility of lung cancer, a variety of data mining approaches, including logistic regression, k-means, and apriori algorithms are used in this study. Age, gender, smoking habit, and medical history are a few of the factors included in the dataset used for the study. The logistic regression classifier shows an accuracy of 95% for the classification of lung cancer in patients. The simulation results obtained from the k-means clustering algorithm shows that the main causes for the possible occurrence of lung cancer are chronic diseases, fatigue, allergy, wheezing, alcohol consumption habit, and breath problem. Similarly, according to the association rule's findings, there is no chance of lung cancer developing in a non-drinker who is free of peer pressure, allergies, and wheezing issues.

Combiner and HyperCombiner networks: Rules to combine multimodality MR images for prostate cancer localisation

One of the distinct characteristics of radiologists reading multiparametric prostate MR scans, using reporting systems like PI-RADS v2.1, is to score individual types of MR modalities, including T2-weighted, diffusion-weighted, and dynamic contrast-enhanced, and then combine these image-modality-specific scores using standardised decision rules to predict the likelihood of clinically significant cancer. This work aims to demonstrate that it is feasible for low-dimensional parametric models to model such decision rules in the proposed Combiner networks, without compromising the accuracy of predicting radiologic labels. First, we demonstrate that either a linear mixture model or a nonlinear stacking model is sufficient to model PI-RADS decision rules for localising prostate cancer. Second, parameters of these combining models are proposed as hyperparameters, weighing independent representations of individual image modalities in the Combiner network training, as opposed to end-to-end modality ensemble. A HyperCombiner network is developed to train a single image segmentation network that can be conditioned on these hyperparameters during inference for much-improved efficiency. Experimental results based on 751 cases from 651 patients compare the proposed rule-modelling approaches with other commonly-adopted end-to-end networks, in this downstream application of automating radiologist labelling on multiparametric MR. By acquiring and interpreting the modality combining rules, specifically the linear-weights or odds ratios associated with individual image modalities, three clinical applications are quantitatively presented and contextualised in the prostate cancer segmentation application, including modality availability assessment, importance quantification and rule discovery.

Significance of Data Structures and Data Retrieval Techniques on Sequence Rule Mining Efficacy

Sequence mining intends to discover rules from diverse datasets by implementing Rule Mining Algorithms with efficient data structures and data retrieval techniques. Traditional algorithms struggle in handling variable support measures which may involve repeated reconstruction of the underlying data structures with changing thresholds. To address these issues the premiere Sequence Mining Algorithm, AprioriAll is implemented against an Educational and a Financial Dataset, using the HASH and the TRIE data structures with scan reduction techniques. Primary idea is to study the impact of data structures and retrieval techniques on the rule mining process in handling diverse datasets. Performance Evaluation Matrices- Support, Confidence and Lifts are considered for testing the efficacies of the algorithm in terms of memory requirements and execution time complexities. Results unveil the excellence of Hashing in tree construction time and memory overhead for fixed sets of pre-defined support thresholds. Whereas, TRIE may avoid reconstruction and is capable of handling dynamic support thresholds, leading to shorter rule discovery time but higher memory consumption. This study highlights the effectiveness of Hash and TRIE data structures considering the dataset characteristics during rule mining. It underscores the importance of appropriate data structures based on dataset features, scanning techniques, and user-defined parameters.

Trading rule discovery using technical analysis and a template matching technique for pattern recognition: Evidence from the Chinese stock market

AbstractThis paper examines the potential profit of bull flag trading rules in the Shangai Stock Exchange Composite Index (SSE) using a template matching technique based on price pattern recognition. This paper fills a gap in the literature by applying a template matching technique for the recognition of bull flag patterns in the Shangai Stock Exchange Composite Index (SSE) during the period of 1991–2021. To the best of our knowledge, no previous study has applied bull flag trading rules to the Chinese stock market. Our results indicate that bull flag trading rules can correctly predict the price movement direction of the index most of the time, achieving significantly positive excess profits. Moreover, shorter fitting windows and better quality of price fit values for lower holding periods are associated with better performance. This research may have relevant practical implications for investors who opt for this indicator in their asset allocation decisions.

Machine learning discovery of optimal quadrature rules for isogeometric analysis

We propose the use of machine learning techniques to find optimal quadrature rules for the construction of stiffness and mass matrices in isogeometric analysis (IGA). We initially consider 1D spline spaces of arbitrary degree spanned over uniform and non-uniform knot sequences, and then the generated optimal rules are used for integration over higher-dimensional spaces using tensor products. The quadrature rule search is posed as an optimization problem and solved by a machine learning strategy based on adaptive gradient-descent. However, since the optimization space is highly non-convex, the success of the search strongly depends on the number of quadrature points and the parameter initialization. Thus, we use a dynamic programming strategy that initializes the parameters from the optimal solution over the spline space with a lower number of knots. With this method, we found optimal quadrature rules for spline spaces when using IGA discretizations with up to 50 uniform elements and polynomial degrees up to 8, showing the generality of the approach in this scenario. For non-uniform partitions, the method also finds an optimal rule in a reasonable number of test cases. We also assess the generated optimal rules in two practical case studies, namely, the eigenvalue problem of the Laplace operator and the eigenfrequency analysis of freeform curved beams, where the latter problem shows the applicability of the method to curved geometries. In particular, the proposed method results in savings with respect to traditional Gaussian integration of up to 44% in 1D, 68% in 2D, and 82% in 3D spaces.