Multi-relational Data Mining Research Articles

Methods for concept analysis and multi-relational data mining: a systematic literature review

Methods for concept analysis and multi-relational data mining: a systematic literature review

Automating Feature Extraction from Entity-Relation Models: Experimental Evaluation of Machine Learning Methods for Relational Learning

With the exponential growth of data, extracting actionable insights becomes resource-intensive. In many organizations, normalized relational databases store a significant portion of this data, where tables are interconnected through some relations. This paper explores relational learning, which involves joining and merging database tables, often normalized in the third normal form. The subsequent processing includes extracting features and utilizing them in machine learning (ML) models. In this paper, we experiment with the propositionalization algorithm (i.e., Wordification) for feature engineering. Next, we compare the algorithms PropDRM and PropStar, which are designed explicitly for multi-relational data mining, to traditional machine learning algorithms. Based on the performed experiments, we concluded that Gradient Boost, compared to PropDRM, achieves similar performance (F1 score, accuracy, and AUC) on multiple datasets. PropStar consistently underperformed on some datasets while being comparable to the other algorithms on others. In summary, the propositionalization algorithm for feature extraction makes it feasible to apply traditional ML algorithms for relational learning directly. In contrast, approaches tailored specifically for relational learning still face challenges in scalability, interpretability, and efficiency. These findings have a practical impact that can help speed up the adoption of machine learning in business contexts where data is stored in relational format without requiring domain-specific feature extraction.

Accelerated and Inexact Soft-Impute for Large-Scale Matrix and Tensor Completion

Matrix and tensor completion aim to recover a low-rank matrix / tensor from limited observations and have been commonly used in applications such as recommender systems and multi-relational data mining. A state-of-the-art matrix completion algorithm is Soft-Impute, which exploits the special "sparse plus low-rank" structure of the matrix iterates to allow efficient SVD in each iteration. Though Soft-Impute is a proximal algorithm, it is generally believed that acceleration destroys the special structure and is thus not useful. In this paper, we show that Soft-Impute can indeed be accelerated without comprising this structure. To further reduce the iteration time complexity, we propose an approximate singular value thresholding scheme based on the power method. Theoretical analysis shows that the proposed algorithm still enjoys the fast $O(1/T^2)$ convergence rate of accelerated proximal algorithms. We further extend the proposed algorithm to tensor completion with the scaled latent nuclear norm regularizer. We show that a similar "sparse plus low-rank" structure also exists, leading to low iteration complexity and fast $O(1/T^2)$ convergence rate. Extensive experiments demonstrate that the proposed algorithm is much faster than Soft-Impute and other state-of-the-art matrix and tensor completion algorithms.

WordificationMI: multi-relational data mining through multiple-instance propositionalization

Multi-relational data mining (MRDM) looks for patterns from a relational database. One of the established approaches to MRDM is propositionalization, characterized by transforming a relational database into a simpler representation, commonly a single table. Another approach that has proven to be effective to address learning problems involving one-to-many relationships between the data is multiple-instance learning. In this paper, we propose a new technique to transform relational data, called WordificationMI, which takes advantage of the multiple-instance learning’s potentialities. This new proposal is based on the bag-of-words representation, proposed in the Wordification methodology, but with the difference that it transforms a relational database into a multiple-instance representation. Additionally, we propose a feature selection method, named MICHI (\(\chi _\mathrm{MI}^{2}\)), for reducing the dimensionality of the datasets obtained with WordificationMI. We also present an empirical evaluation with ten relational databases and four learning techniques that show the effectiveness of the proposed methods.

A User-Driven Association Rule Mining Based on Templates for Multi-Relational Data

Data mining algorithms to find association rules are an important tool to extract knowledge from databases. However, these algorithms produce an enormous amount of rules, many of which could be redundant or irrelevant for a specific decision-making process. Also, the use of previous knowledge and hypothesis are not considered by these algorithms. On the other hand, most existing data mining approaches look for patterns in a single data table, ignoring the relations presented in relational databases. The contribution of this paper is the proposition of a multi-relational data mining algorithm based on association rules, called TBMR-Radix, which considers previous knowledge and hypothesis through the using of the Templates technique. Applying this approach over two real databases, we were able to reduce the number of generated rules, use the existing knowledge about the data and reduce the waste of computational resources while processing. Our experiments show that the developed algorithm was also able to perform in a multi-relational environment, while the MR-Radix, that does not use Templates technique, was not.

Evolutionary Propositionalization of Multi-Relational Data — Research Notes

Propositionalization has been proven to be a very effective solution for multi-relational data mining problems. The approaches usually follow a two-step principle: transforming the relational data into a single, flat table and applying a propositional learning algorithm. During the transformation, the target table gets expanded by adding many new features summarizing the information of the non-target tables. Based on the used feature construction strategy, this leads to a table of very high dimensionality with a lot of irrelevant and/or redundant features that can negatively affect the predictive performance. In this paper, we propose a modification of the traditional two-step framework to overcome such problems. The proposed approach evaluates the features during the construction phase and reports only a subset of highly predictive features to the propositional learner. We present an implementation of this approach using a genetic algorithm to search for an optimal feature subset. Our experiments on a number of benchmark datasets suggest that the modified framework can help propositionalization methods to significantly improve their predictive performance.

Parallel algorithms for enumerating closed patterns from multi-relational data

This paper presents parallel algorithms for enumerating closed patterns from multi-relational data. In multi-relational data mining (MRDM), patterns are represented in logical formulae, and involve multiple tables (relations) from a relational database. Since the expressive framework of MRDM makes the task of pattern mining costly compared with the conventional itemset mining, we propose parallel algorithms for computing closed patterns on multi-core processors. In particular, we present new load-balancing strategies which try to fully exploit the task-parallelism intrinsic in the search process of the problem, and give some experimental results, which show the effectiveness of the proposed methods. We then apply our proposed methods to compute closed patterns, a.k.a. concept intents, for binary object-attribute relational data, and show by experiments that the performance of our method is comparable to the existing method.

Using multi-relational data mining to discriminate blended therapy efficiency on patients based on log data

IntroductionClinical trials of blended Internet-based treatments deliver a wealth of data from various sources, such as self-report questionnaires, diagnostic interviews, treatment platform log files and Ecological Momentary Assessments (EMA). Mining these complex data for clinically relevant patterns is a daunting task for which no definitive best method exists. In this paper, we explore the expressive power of the multi-relational Inductive Logic Programming (ILP) data mining approach, using combined trial data of the EU E-COMPARED depression trial. MethodsWe explored the capability of ILP to handle and combine (implicit) multiple relationships in the E-COMPARED data. This data set has the following features that favor ILP analysis: 1) Time reasoning is involved; 2) there is a reasonable amount of explicit useful relations to be analyzed; 3) ILP is capable of building comprehensible models that might be perceived as putative explanations by domain experts; 4) both numerical and statistical models may coexist within ILP models if necessary. In our analyses, we focused on scores of the PHQ-8 self-report questionnaire (which taps depressive symptom severity), and on EMA of mood and various other clinically relevant factors. Both measures were administered during treatment, which lasted between 9 to 16 weeks. ResultsE-COMPARED trial data revealed different individual improvement patterns: PHQ-8 scores suggested that some individuals improved quickly during the first weeks of the treatment, while others improved at a (much) slower pace, or not at all. Combining self-reported Ecological Momentary Assessments (EMA), PHQ-8 scores and log data about the usage of the ICT4D platform in the context of blended care, we set out to unveil possible causes for these different trajectories. DiscussionThis work complements other studies into alternative data mining approaches to E-COMPARED trial data analysis, which are all aimed to identify clinically meaningful predictors of system use and treatment outcome. Strengths and limitations of the ILP approach given this objective will be discussed.

An efficient multi relational framework using fuzzy rule-based classification technique

Multi relational classification is one of the rapidly rising subclasses of multi relational data mining. It focuses on identifying attractive patterns in an interconnected multiple database relations via primary key/foreign key relationship. Many conventional rule-based classification approaches produce the rules from single 'flat' relation and typically suffer from rule clashes. However, many real world data repositories are stored in multiple database relations. Owing to the increase in the quantity of relational data that are being accumulated and the limitation of propositional rule-based classifiers in relational domains, multi relational classification has turned out to be a field with significance. In this work, a multi relational fuzzy rule system (MRFRS) is proposed for classification across multiple database relations based on PART algorithm that generates crisp rules. These rules are used to formulate the multi relational fuzzy rule system that produces rules of mutually exclusive and exhaustive in nature. The empirical results show that the proposed method outperforms well on real datasets in terms of efficiency and accuracy.

NHL and RCGA Based Multi-Relational Fuzzy Cognitive Map Modeling for Complex Systems

In order to model multi-dimensions and multi-granularities oriented complex systems, this paper firstly proposes a kind of multi-relational Fuzzy Cognitive Map (FCM) to simulate the multi-relational system and its auto construct algorithm integrating Nonlinear Hebbian Learning (NHL) and Real Code Genetic Algorithm (RCGA). The multi-relational FCM fits to model the complex system with multi-dimensions and multi-granularities. The auto construct algorithm can learn the multi-relational FCM from multi-relational data resources to eliminate human intervention. The Multi-Relational Data Mining (MRDM) algorithm integrates multi-instance oriented NHL and RCGA of FCM. NHL is extended to mine the causal relationships between coarse-granularity concept and its fined-granularity concepts driven by multi-instances in the multi-relational system. RCGA is used to establish high-quality high-level FCM driven by data. The multi-relational FCM and the integrating algorithm have been applied in complex system of Mutagenesis. The experiment demonstrates not only that they get better classification accuracy, but it also shows the causal relationships among the concepts of the system.

Distributed Mining of Closed Patterns from Multi-Relational Data

In multi-relational data mining (MRDM), there have been proposed many methods for searching for patterns that involve multiple tables (relations) from a relational database. In this paper, we consider closed pattern mining from distributed multi-relational databases (MRDBs). Since the computation of MRDM is costly compared with the conventional itemset mining, we propose some efficient methods for computing closed patterns using the techniques studied in Inductive Logic Programming (ILP) and Formal Concept Analysis (FCA). Given a set oflocaldatabases, we first compute sets of their closed patterns (concepts) using a closed pattern mining algorithm tailored to MRDM, and then generate the set of closed patterns in the global database by utilizing themergeoperator. We also present some experimental results, which shows the effectiveness of the proposed methods.

Graph Representation of Relational Database for Concept Discovery

Multi-relational concept discovery aims to find the relational rules that best describe the target concept. In this paper, we present a graph-based concept discovery method in Multi-Relational Data Mining. Concept rule discovery aims at finding the definition of a specific concept in terms of relations involving background knowledge. The proposed method is an improvement over a state-of-the-art concept discovery system that uses both ILP and conventional association rule mining techniques during concept discovery process. The proposed method generates graph structures with respect to data that is initially stored in a relational database and utilizes them to guide the concept induction process. A set of experiments is conducted on data sets that belong to different learning problems. The results show that the proposed method has promising results in comparison to state of the art methods.

Entropy Based Feature Selection For Multi-Relational Naïve Bayesian Classifier

INTRODUCTION The term data mining refers to the extraction of valuable knowledge from large amounts of data. Data mining is the process of discovering knowledge from data. With the massive quantity of data stored in repositories, it is progressively more significant to develop powerful analysis and decision making tool for the extraction of interesting knowledge. The task of classification is concerned with predicting the value of one field from the values of other field. The target field is called the class. The other fields are called attributes. Propositional machine learning algorithms assume the input data is represented in a simple attribute-value format. Most existing data mining algorithms (including algorithms for classification, clustering, association analysis, outlier detection, etc.) work on single tables. For example, a typical classification algorithm (e.g., C4.5 or SVM) works on a table containing many tuples, each of which has a class label, and a value on each attribute in the table. In recent years, there has been growing interest in multi-relational classification research and application, which address the difficulties in dealing with large relation search space, complex relationships between relations, and a daunting number of attributes involved. Most structured data is stored in relational databases, which is stored in multiple relations by their characters. Conventionally, many classification approaches can only be applied to a single relation. When performing these approaches on multi-relational data, it often requires transferring data into a single table by flattening and feature construction, which is known as Propositionalization. However, many of these methods are heuristic, so flatten may cause some problems such as time consuming and statistical skew on data. Multi-relational data mining (MRDM) has been successfully applied in a variety of areas, such as marketing, sales, finance, fraud detection, and natural sciences. Multi-Relational data mining looks for patterns that involve multiple relations in a relational database, its main difference with traditional data mining approaches is that it does not need to transform the data into a single table, it learns from the data in its original form preserving its structure and incorporating such structure into the learning process. RELATIONAL DATABASES A relational database is a collection of tables called relations, each of which is assign a unique name. Each relation consists of a set of attributes and stores a large set of tuples. Every tuple in a relational table represents an object which is used to identifying by a unique key to describe by a set of attribute values. Often one uses a semantic model to represent relational databases, allowing one to describe and design the database without having to pay attention to the physical database. Such a model is often referred to as a database scheme. One of the most common models is the Entity-Relationship (ER) model (Figure 1). A relational database typically consists of several tables (relations) and not just one table. A schema for a relational databases describe a set of entities DB = {E1, E2, En}, and set of relationships between entities. Each row in a relation is a tuple. Each relation has at least one primary key attributes. The other attributes are either descriptive attributes or foreign key attributes. Foreign key attributes link to primary key attribute of other relations. A relational database contains multiple interconnected relations, each of which represents a certain kind of objects or a type of relationships. A relational database consists of a set of named tables, often referred to as relations that individually behave as the single table that is the subject of Propositional Data Mining. Data structures more complex than a single record are implemented by relating pairs of tables through so-called foreign key relations. Such a relation specifies how certain columns in one table can be used to look up information in corresponding columns in the other table, thus relating sets of records in the two tables. …

Information Theory Based Feature Selection for Multi-Relational Naïve Bayesian Classifier

Today data’s are stored in relation structures. In usual approach to mine these data, we often use to join several relations to form a single relation using foreign key links, which is known as flatten. Flatten may cause troubles such as time consuming, data redundancy and statistical skew on data. Hence, the critical issues arise that how to mine data directly on numerous relations. The solution of the given issue is the approach called multi-relational data mining (MRDM). Other issues are irrelevant or redundant attributes in a relation may not make contribution to classification accuracy. Thus, feature selection is an essential data pre-processing step in multi-relational data mining. By filtering out irrelevant or redundant features from relations for data mining, we improve classification accuracy, achieve good time performance, and improve comprehensibility of the models. We had proposed the entropy based feature selection method for Multi-relational Naive Bayesian Classifier. We have use method InfoDist and Pearson’s Correlation parameters, which will be used to filter out irrelevant and redundant features from the multi-relational database and will enhance classification accuracy. We analyzed our algorithm over PKDD financial dataset and achieved the better accuracy compare to the existing features selection methods.

Genetic algorithm-based optimized association rule mining for multi-relational data

The multi relational data mining is one of the latest topics in data mining to find the relational patterns. In this paper, we have presented an algorithm for multi-relational rule mining using association rule mining and the optimization process. As a result of the association rule mining on the multirelational data, a number of relevant and irrelevant rules are generated. A rule is specified as a relation between two data points in the dataset. So, an optimization should be done on the mining algorithm in order to get the most relevant rules. We have adapted the technique of genetic algorithm in order to optimize the mined multi relational association rules. The genetic algorithm is one of the best optimization algorithm available and it suites the current problem because of its particular features such as the genetic operators crossover and mutation. The optimization of the rule is done by altering the fitness function of the genetic algorithm in relation with the multi relational data mining algorithm. The results from the experimental analysis showed that the proposed approach has better efficiency over the previous approaches. The most rules optimized is 198 under iterations 10 with a support of 60.

New Challenges for Clustering in Large Data Base

analysis in data mining is a main application of business. This Investigation describes to present NCDBC algorithm that extends expansion seed selection into a DBSCAN algorithm. And the DBSCAN Algorithm describes the density based clustering concept and also describes its hierarchical additional room OPTICS has been planned newly, and one of the mainly triumphant approaches to clustering. Aim of this research work is to move on the high- tech clustering; mainly density-based clustering by identifying new challenges for density based clustering and proposing inventive for these challenges. In this work the proposed procedure focuses on decrease the number of seeds points and also reduces the execution time cost of searching neighborhood data. And A hierarchical clustering procedure can be useful to these interesting subspaces in order to calculate a Latitude for north and south cities and also calculate Longitude of different cities. The relational record theory select data concentrated applications of engineering level. the Relational data base used in this application for data storage and recover. The main reason of the use of relational data base will provide encouraging practical come near on data of Data Mining. This database presumption has a absolute and well-off account of opinion and developments concerning to the well-organized storage and dealing out of prepared data with extensive knowledge of Multi-Relational Data Mining. data mining Concepts which emphasis to provide a model for a data base and normalize that data base will provide the help for NCDBCLE plan. there are many of the Current developments which deal with large volume of data base and use of concentrated processing for data base provide improved application of NCDBCLE in more finer and complicated areas. In the relational database theory there are various concepts and different customs to encouraged other prepared Data Mining paradigms. NCDBCLE fundamentals various approaches for larger records. the proposed work provide the understandability for relational groundwork. which is helpful for the concept of this proposed work, the proposed work successfully created solution that is not highlighted before in any challenges in this area. 2. REVIEW

A Technique to Mine the Multi-Relational Patterns Using Relational Tree and a Tree Pattern Mining Algorithm

The reason of the powerful application and large availability of the databases, the data mining become chalengable and trustful research field. The pattern mining is one of the fields in the data mining. The multi-relational data mining allows pattern mining from multiple table, in recent years multi-relational pattern mining is developing rapid manner. Even though the existing multi-relational mining algorithms are not ideal for the large amount of data. In this paper, we have developed a technique to mine the multi-realtional pattern using a relational tree. The relational tree constructed by reading the records from multi-relational database one by one and make the combination (relations) according the presence of fields in the database. From this we get the relational tree data structure without violating the relations of the records. Subsequently , the tree pattern mining algorithm is formulated and applied to the developed tree structure for mining the important relational patterns from the relational tree. The experimentation is carried out on the patient medical database and comparative results are extracted and the performance of the proposed approach is evaluated based on the existing work.

State of Art of Multi Relational Data Mining Approaches: A Rule Mining Algorithm

In this 21 st century is completely called as the information science where the large organizations need useful knowledge. The data mining algorithms look for patterns in data. While most existing data mining approaches look for patterns in a single data table, multi-relational data mining (MRDM) approaches look for patterns that involve multiple tables (relations) from a relational database. The database consists of a collection of tables (a relational database). Records in each table represent parts, and individuals can be reconstructed by joining over the foreign key relations between the tables. To reduce the I/O cost, the data accessed together during extraction phase are to be clustered in the same disk block. This paper represents the index structure ,what we generally called as the Imine index structure .This structure can efficiently exploited by different item set extraction as well as this novel index structure is implemented by using FPGrowth and LCM V.2 algorithms. Again in this paper we have focused that how the MRDM techniques are used in different approaches like classification, clustering ILP (Inductive Logic Program) etc.

A Review: Data mining over Multi-Relations

In this paper, Multi-relational data mining enables pattern mining from multiple tables. Multi-relational data mining algorithms can be used as practical proposal to overcome the deficiency of conventional algorithms. Multi-relational data mining algorithms directly extract frequent patterns from different registers in efficient manner without need of transfer the data in a single table will, on the other hand, used the available memory space is not enough to ensure the production of large amounts of data. For this reason, and the use of space, algorithms are an integral care for the prospection of large repositories. The paper provides the overview of multi relation data mining techniques and classification algorithms. It also defines the frequent pattern mining. The presented paper discussed the various architecture and issues related to multi table data mining. A lot of literature has been proposed in this area. Some of them has discussed in this paper.

Relational concept analysis: mining concept lattices from multi-relational data

The processing of complex data is admittedly among the major concerns of knowledge discovery from data (kdd). Indeed, a major part of the data worth analyzing is stored in relational databases and, since recently, on the Web of Data. This clearly underscores the need for Entity-Relationship and rdf compliant data mining (dm) tools. We are studying an approach to the underlying multi-relational data mining (mrdm) problem, which relies on formal concept analysis (fca) as a framework for clustering and classification. Our relational concept analysis (rca) extends fca to the processing of multi-relational datasets, i.e., with multiple sorts of individuals, each provided with its own set of attributes, and relationships among those. Given such a dataset, rca constructs a set of concept lattices, one per object sort, through an iterative analysis process that is bound towards a fixed-point. In doing that, it abstracts the links between objects into attributes akin to role restrictions from description logics (dls). We address here key aspects of the iterative calculation such as evolution in data description along the iterations and process termination. We describe implementations of rca and list applications to problems from software and knowledge engineering.