An Integration Method Of Classifiers For Abnormal Phone Detection

Classification and associative rule mining are two substantial areas in data mining. Some scientists attempt to integrate these two field called rule-based classifiers. Rule-based classifiers can play a very important role in applications such as fraud detection, medical diagnosis, and etc. Numerous previous studies have shown that this type of classifiers achieves high classification accuracy than traditional classification algorithms. However, they still suffer from a fundamental limitation. Many rule-based classifiers used various greedy techniques to prune the redundant rules that lead to missing some important rules. Another challenge that must be considered is related to the enormous set of mined rules that result in high processing overhead. The result of these approaches is that the final selected rules may not be the global best rules. These algorithms are not successful at exploiting search space effectively in order to select the best subset of candidate rules. We merged the Apriori algorithm, harmony search, and classification based association rules (CBA) algorithm for building a rule-based classifier. We applied a modified version of the Apriori algorithm with multiple minimum support for extracting useful rules for each class in the dataset. Instead of using a large number of candidate rules, binary harmony search was utilized for selecting the best subset of rules that appropriate for building a classification model. We applied the proposed method on seventeen benchmark dataset and compared its result with traditional association rule classification algorithms. The statistical results show that our proposed method outperformed other rule-based approaches.

Classification and associative rule mining are two substantial areas in data mining. Some scientists attempt to integrate these two field called rule-based classifiers. Rule-based classifiers can play a very important role in applications such as fraud detection, medical diagnosis, etc. Numerous previous studies have shown that this type of classifier achieves a higher classification accuracy than traditional classification algorithms. However, they still suffer from a fundamental limitation. Many rule-based classifiers used various greedy techniques to prune the redundant rules that lead to missing some important rules. Another challenge that must be considered is related to the enormous set of mined rules that result in high processing overhead. The result of these approaches is that the final selected rules may not be the global best rules. These algorithms are not successful at exploiting search space effectively in order to select the best subset of candidate rules. We merged the Apriori algorithm, Harmony Search, and classification-based association rules (CBA) algorithm in order to build a rule-based classifier. We applied a modified version of the Apriori algorithm with multiple minimum support for extracting useful rules for each class in the dataset. Instead of using a large number of candidate rules, binary Harmony Search was utilized for selecting the best subset of rules that appropriate for building a classification model. We applied the proposed method on a seventeen benchmark dataset and compared its result with traditional association rule classification algorithms. The statistical results show that our proposed method outperformed other rule-based approaches.

Classifying bare earth (ground) points from Light Detection and Ranging (LiDAR) point clouds is well-established research in the forestry, topography, and urban domains using point clouds acquired by Airborne LiDAR System (ALS) at average point densities (≈2 points per meter-square (pts/m2)). The paradigm of point cloud collection has shifted with the advent of unmanned aerial systems (UAS) onboard affordable laser scanners with commercial utility (e.g., DJI Zenmuse L1 sensor) and unprecedented repeatability of UAS-LiDAR surveys. Therefore, there is an immediate need to investigate the existing methods, and to develop new ground classification methods, using UAS-LiDAR. In this paper, for the first time, traditional ground classification algorithms and modern machine learning methods were investigated to filter ground from point clouds of high-density UAS-LiDAR data (≈900 pts/m2) over five agricultural fields in North Dakota, USA. To this end, we tested frequently used ground classification algorithms: Cloth Simulation Function (CSF), Progressive Morphological Filter (PMF), Multiscale Curvature Classification (MCC), and ArcGIS ground classification algorithms along with the PointCNN deep learning model were trained. We investigated two aspects of ground classification algorithms and PointCNN: (a) Classification accuracy of optimized ground classification algorithms (i.e., fine adjustment is user-defined parameters) and PointCNN over training site, and (b) transferability potential over four yet diverse test agricultural fields. The well-established evaluation metrics of omission error, commission error, and total error, along with kappa coefficients showed that deep learning outperforms the traditional ground classification algorithms in both aspects: (a) overall classification accuracy, and (b) transferability over diverse agricultural fields.

With the rapid development of the Internet of Things, satellite communications, unmanned aerial vehicles, and other technologies, the application of data processing technology in the field of emergency rescue has attracted great attention. In this paper, a cleaning algorithm for Space-Air-Ground integrated rescue data is proposed, which can be used to solve the problems caused by dirty data, such as the long data processing time and poor rescue efficiency. Based on the traditional sorted neighborhood method (SNM), a jump-sliding window strategy is added to the proposed cleaning algorithm. By reducing the data comparisons, the proposed cleaning algorithm has greatly shortened the cleaning time of duplicate approximate records in the rescue data sets, while satisfying the data quality. The rescue data with different scales are used to test the two algorithms, and the simulation results show that the processing time of the proposed algorithm is far less than that of the traditional SNM algorithm. For example, when the data scale is 1.827 million and the window size is 50, the processing time of the proposed algorithm is 27s, which is 10 times shorter than that of the traditional SNM algorithm. When the window size changes to 500, the processing time of the proposed algorithm is 44s, which is 28.7 times shorter than that of the traditional SNM algorithm. So the proposed algorithm is more suitable for cleaning duplicate approximate rescue data, due to its urgent time requirement for rescue events.

Data is the new Gold of the modern age. Organizational decisions, artificial intelligence, machine learning, time critical processes, analytics, etc. are driven by highly profiled data. Data gives an organization the ability to act in a corresponding situation and hence, keep up with the market. Financial returns are directly associated with the quality of data being used. Today’s world has experienced a data explosion with the advent of data sources like Hadoop, data lakes, clusters and many more. Even more, the day to day devices that we use collect data to present meaningful statistics to us. There is an exponential increase in data volumes which are characterized by heterogeneous data formats and disparate data sources which in turn is complicating the task of providing uniform and profiled data to the business users for analysis. Warehousing is also not able to cope up with the on-demand real-time data access. Holistic view of the entire data is extremely difficult to achieve and takes a considerable amount of time which is slowing down an organization’s decision making capabilities. The complexity of the current systems has increased due to the interconnected nature of widely distributed systems which share data. Dirty data i.e. data which is inaccurate, inconsistent, unclean or incomplete is spread all across. As all the data is unrequired garbage data, there is no point in developing your business intelligence over this data. Data Clarity used along with Data Virtualization provides an exceptional solution to these problems by handling the complex data workloads with respect to heterogeneous data and disparate data sources and provides a holistic view of the data to the business organization.

McGraw-Hill Education’s new adaptive flashcard application, StudyWise, implements spaced practice to help learners memorize collections of basic facts. For classroom use, subject matter experts needed a scheduling algorithm that could provide effective practice schedules to learn a pre-set number of facts over a specific interval of days. To test the pedagogical effectiveness of such schedules, we used the ACT-R model of memorization to simulate learner responses. Each schedule has one 30 min study session per day, with overall study intervals that ranged from one day for sets of less than 30 items to three weeks for sets of two hundred or more items. In each case, we succeeded in tuning our algorithm to give a high probability the simulated learner answered each item correctly by the end of the schedule. This use of artificial intelligence allowed us to optimize the algorithm before engaging large numbers of real users. As real user data becomes available for this application, the simulated user model can be further tested and refined.

Web applications are crucial role for daily user activities such as online banking, online shopping and searching. It is important to ensure the reliability and web application testing has been used in finding various faults in order to improve the quality of reliable web services. Among test cases generation approaches, user session based testing is an approach to create test cases with real user data. However, real user data usage is extremely large and executing all the test cases can be time consuming in practice. This paper describes the test cases reduction approach for analyzing and replaying the large number of test cases generated from user session data. The entropy gain theory is applied in test cases reduction process to get the best test suite that covers all user accesses of web application. To evaluate the effectiveness of proposed method, the analytical results are described in terms of URLs coverage, reduction time and test cases reduction rate.

This paper put forward a classification model of Tibetan medical syndrome of common plateau disease (Chronic Atrophic Gastritis) based on the class association rules. Taking the clinical diagnosis and treatment data of Chronic Atrophic Gastritis provided by Qinghai Provincial Tibetan Hospital as the research object. Firstly, using the clustering algorithm to divide syndrome normalized. Secondly, the constraint-based Apriori algorithm is applied to mine the strong association rules between symptoms and syndrome after pruning and prioritizing these rules, and the classification model of the Tibetan medical syndrome is established by selecting a set with high-priority rules. Finally, the accuracy of the classification model is verified and compared with four traditional classification algorithms to prove the effectiveness of the proposed method. The experimental results showed that the proposed method outperformed most other traditional classification algorithms with an accuracy of 82.72%.

There are many traditional classification algorithms used to classify multispectral images, especially those used in remote sensing. But the challenges of using these algorithms for multispectral image classification are that they are slow to implement and have poor classification accuracy. With the development of technologies that mimic nature, many researchers have resorted to using intelligent algorithms instead of traditional algorithms because of their great importance, especially when dealing with large amounts of data. The bat algorithm (BA) is one of the most important of these algorithms. This study aims to verify the possibility of using the BA to classify the multispectral images captured by the Landsat-5 TM satellite image of the study area. The study area represents the Mosul area located in the Nineveh Governorate in northwestern Iraq. The purpose is not only to study the ability of the BA to classify multispectral images but also to obtain a land cover map of this region. The BA showed efficiency in the classification results compared to Maximum Likelihood (ML), where the overall accuracy of classification when using the BA reached (82.136%), while MLreached (79.64%)

In recent years, data quality issues have attracted wide attention. Data quality is mainly caused by dirty data. Currently, many methods for dirty data management have been proposed, and one of them is entity-based relational database in which one tuple represents an entity. The traditional query optimizations having the ability to estimate the cost of execution of a query plan have not been suitable for the new entity-based model. Then new query optimizations need to be developed. In this paper, we propose new query selectivity estimation based on histogram, and focus on solving the overestimation which traditional methods lead to. We prove our approaches are unbiased. The experimental results on both real and synthetic data sets show that our approaches can give good estimates with low error.Keywordsquery estimationhistogramdirty datadata quality

This chapter focuses on the process of cleaning data and preparing it for further processing. Specifically, the chapter discusses various techniques that you will use, including preprocessing, outlier identification, data consistency, and the normalization or standardization process, used to normalize your data. The chapter further discusses different measurement types and what methods can be used for which types. The chapter also discusses ways to deal with issues you may encounter with inconsistent or dirty data. The chapter takes a more practical approach by integrating several labs with actual game data to demonstrate how you can perform these steps on real game data.

Large language models (LLMs) are increasingly used for mental health, yet safety evaluations rely primarily on small, simulation-based benchmarks removed from real-world language. We replicate four published safety evaluations assessing suicide risk handling, harmful content generation, and jailbreak resistance for general-purpose frontier models and a purpose-built mental health AI. We then conduct an ecological audit of 20,000 real user conversations with the purpose-built system, which includes layered safeguards for suicide and non-suicidal self-injury (NSSI). The purpose-built AI was significantly less likely than general-purpose LLMs to produce harmful content across suicide/NSSI (.4-11.27% vs 29.0-54.4%), eating disorder (8.4% vs 54.0%), and substance use (9.9% vs 45.0%) benchmarks. In real user data, clinician review found zero suicide-risk cases without crisis resources. Three NSSI mentions (.015%) lacked intervention, implying a .38% lower-bound false negative rate. Findings support the utility of ecological audits for safety estimation.

Detection of credit card fraud has lately been considered a critical task due to the highly imbalanced nature of financial transaction databases. On the other hand, the traditional classification algorithms have been poor at detecting fraudulent activities with an acceptable false-positive rate. Hence, this work contributes to a GMM based approach for fraud detection, which benefits from GMM probabilistic-based classification power for better classification results. The database used in this study is publicly available and was obtained from the Kaggle Credit Card Fraud Detection (CCFD) database. The dataset has 284,807 transactions, and only 0.17% of the cases represent fraud. This paper plans to scale the features, train the GMM technique with different numbers of Gaussian components (i.e., 2, 4, 6, 8, and 10), and evaluate their performance with several evaluation metrics. Compared with other traditional classifiers (logistic regression (92.4%), K-nearest neighbors (93.68%), decision tree (88.16%), and support vector machine (94.21%)), the proposed GMM algorithm obtains the highest accuracy of 94.53%. The proposed method, despite its high accuracy, has limitations under high-dimensional feature dependencies and optimal component selection. From the results obtained over the experimentation process, the GMM proves to be a probable, yet flexible and subservient framework for a complex modelling of probabilities for the detection of fraud.

An application of a new meta-heuristic for optimizing the classification accuracy when analyzing some medical datasets

This paper will investigate the effectiveness of the genetic algorithm evolved neural network classifier and its application on the land cover classification of multi-spectral remotely sensed imagery. First, the key issues of the algorithms and the procedures are described in detail. Second, SPOT XS imagery is employed to evaluate its accuracy. Traditional classification algorithms, such as maximum likelihood classifier, back propagation neural network classifier, are also incorporated for a comparison purpose. Based on an evaluation of the user's accuracy and kappa statistic of different classifiers, the superiority of applying the discussed genetic algorithm-based classifier for land cover classification using multi-spectral imagery data is established. Finally, some concluding remarks and suggestions are also presented.