K-nearest Neighbor Technique Research Articles

Can the MMPI Predict Adult ADHD? An Approach Using Machine Learning Methods.

(1) Background: Adult attention-deficit/hyperactivity disorder (ADHD) symptoms cause various social difficulties due to attention deficit and impulsivity. In addition, in contrast to ADHD in childhood, ADHD in adulthood is difficult to diagnose due to mixed psychopathologies. This study aimed to determine whether it is possible to predict ADHD symptoms in adults using the Minnesota Multiphasic Personality Inventory-2 (MMPI-2) with machine learning (ML) techniques; (2) Methods: Data collected from 5726 college students were analyzed. The MMPI-2-Restructured Form (MMPI-2-RF) was used, and ADHD symptoms in adults were evaluated using the Attention-Deficit/Hyperactivity Disorder Self-Report Scale (ASRS). For statistical analysis, three ML algorithms were used, i.e., K-nearest neighbors (KNN), linear discriminant analysis (LDA), and random forest, with the ASRS evaluation result as the dependent variable and the 50 MMPI-2-RF scales as predictors; (3) Results: When the KNN, LDA, and random forest techniques were applied, the accuracy was 93.1%, 91.2%, and 93.6%, respectively, and the area under the curve (AUC) was 0.722, 0.806, and 0.790, respectively. The AUC of the LDA method was the largest, with an excellent level of diagnostic accuracy; (4) Conclusions: ML using the MMPI-2 in a large group could provide reliable accuracy in screening for adult ADHD.

Microgrids islanding detection using Fourier transform and machine learning algorithm

This paper proposes a methodology to identify the decision of microgrid islanding. This methodology is two stages: feature extraction and classification stages. In the feature extraction, instantaneous three phase voltage and current measurements are processed using discrete Fourier transform and the symmetrical components of voltage, current and the voltage times current of the second order harmonic are extracted. These nine extracted features are inputs to the second stage that is based on a machine learning algorithm called K-nearest neighbor technique (KNN). In KNN, a small part of the processed data is trained to get the predictor model that is used to the residual data to identify the islanding decision. A MATLAB simulation model is implemented to simulate the intentional and unintentional islanding events and a lab experiment is set up to make these islanding events practically. In addition, the impact of different types of distributed generation are studied to assess their impacts on the performance of the proposed methodology. Simulation and experimental results show that the proposed methodology has the ability to identify the correct islanding decision with the presence of noise with high accuracy and small detection time. In addition, the methodology has the capability to distinguish between islanding and non-islanding. The comparison with other algorithms shows a higher accuracy, dependability, security and a smallest islanding detection time.

Social & Juristic challenges of AI for Opinion Mining Approaches on Amazon & Flipkart Product Reviews Using Machine Learning Algorithms

In recent years, the retail market industry has taken a broad form to sell the products online and also to give the opportunity to customers to provide their valuable feedbacks, suggestions and recommendations. The opinion summarization and classification systems extract and identify a range of opinions about different online available products in a large text-based review set. This paper addresses and reviews the concepts of automatic identification of the sentiments expressed in the English text for Amazon and Flipkart products using Naive Bayes, Logistic Regression, SentiWordNet, Random Forest and K-Nearest Neighbor techniques. It presents a detailed comparative study of such existing sentiment analysis algorithms and methodologies on the basis of five key parameters. The paper also proposed Product Comment Summarizer and Analyzer (PCSA) system. It is automatic and generic comment analyzer which can find the polarity of the sentiments and comments very effectively. It summarizes the comments and classifies them into the pre-defined positive, negative or neutral classes. It results in evaluating their performance in terms of parameter rating, classifiers, accuracy.

Classification of Agriculture Farm Machinery Using Machine Learning and Internet of Things

In this paper, we apply the multi-class supervised machine learning techniques for classifying the agriculture farm machinery. The classification of farm machinery is important when performing the automatic authentication of field activity in a remote setup. In the absence of a sound machine recognition system, there is every possibility of a fraudulent activity taking place. To address this need, we classify the machinery using five machine learning techniques—K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF) and Gradient Boosting (GB). For training of the model, we use the vibration and tilt of machinery. The vibration and tilt of machinery are recorded using the accelerometer and gyroscope sensors, respectively. The machinery included the leveler, rotavator and cultivator. The preliminary analysis on the collected data revealed that the farm machinery (when in operation) showed big variations in vibration and tilt, but observed similar means. Additionally, the accuracies of vibration-based and tilt-based classifications of farm machinery show good accuracy when used alone (with vibration showing slightly better numbers than the tilt). However, the accuracies improve further when both (the tilt and vibration) are used together. Furthermore, all five machine learning algorithms used for classification have an accuracy of more than 82%, but random forest was the best performing. The gradient boosting and random forest show slight over-fitting (about 9%), but both algorithms produce high testing accuracy. In terms of execution time, the decision tree takes the least time to train, while the gradient boosting takes the most time.

MODELLING AND PREDICTING MEASURES OF TREE SPECIES DIVERSITY USING AIRBORNE LASER SCANNING DATA IN MIOMBO WOODLANDS OF TANZANIA

In the recent decade, remote sensing techniques had emerged as one among the best options for quantification of measures of tree species diversity. In this study, potential of using remotely sensed data derived from airborne laser scanning (ALS) for predicting tree species richness and Shannon diversity index was evaluated. Two modelling approaches were tested: linear mixed effects modelling (LMM), by which each of the measures was modelled separately, and the k-nearest neighbour technique (k-NN), by which both measures were jointly modelled (multivariate approach). For both methods, the effect of vegetation type on the prediction accuracies of tree species richness and Shannon diversity index was tested. Separate predictions for richness and Shannon diversity index using LMM resulted in relative root mean square errors (RMSEcv) of 40.7%, and 39.1%, while for the k-NN they were 41.4% and 39.1%, respectively. Inclusion of dummy variables representing vegetation types to the LMM improved the prediction accuracies of tree species richness (RMSEcv = 40.2%) and Shannon diversity index (RMSEcv = 38.0%). The study concluded that ALS data has a potential for modelling and predicting measures of tree species diversity in the miombo woodlands of Tanzania.

An Analysis of a KNN Perturbation Operator: An Application to the Binarization of Continuous Metaheuristics

The optimization methods and, in particular, metaheuristics must be constantly improved to reduce execution times, improve the results, and thus be able to address broader instances. In particular, addressing combinatorial optimization problems is critical in the areas of operational research and engineering. In this work, a perturbation operator is proposed which uses the k-nearest neighbors technique, and this is studied with the aim of improving the diversification and intensification properties of metaheuristic algorithms in their binary version. Random operators are designed to study the contribution of the perturbation operator. To verify the proposal, large instances of the well-known set covering problem are studied. Box plots, convergence charts, and the Wilcoxon statistical test are used to determine the operator contribution. Furthermore, a comparison is made using metaheuristic techniques that use general binarization mechanisms such as transfer functions or db-scan as binarization methods. The results obtained indicate that the KNN perturbation operator improves significantly the results.

A KNN quantum cuckoo search algorithm applied to the multidimensional knapsack problem

Optimization algorithms and particularly metaheuristics are constantly improved with the goal of reducing execution times, increasing the quality of solutions, and addressing larger target instances. Hybridizing techniques are one of these methods particularly interesting for the broad scope of problems to which they can be adapted. In this work, we assessed a hybrid algorithm that uses the k-nearest neighbor technique to improve the results of a quantum cuckoo search algorithm for resource allocation. The k-nearest neighbor technique is used to direct the movement of solutions. Numerical experiments were performed to obtain insights from the contribution of the k-nearest neighbor technique in the final result of solutions. The well-known multidimensional knapsack problem was addressed in order to validate our procedure; a comparison is made with state-of-the-art algorithms. Our results show that our hybrid algorithm consistently produces better results in most of the analyzed instances.

Prediction of Succinic Acid Extraction Efficiency in the Emulsion Liquid Membrane by using Machine Learning Techniques

This research aims to predict succinic acid concentration in the external phase during the emulsion liquid membrane process by using artificial neural networks along with a popular alternative method: k-nearest neighbor technique. The solute concentration values can be predicted by the proposed method without performing a great number of emulsion liquid membrane experiments. Several computer simulations were performed to demonstrate the success of the system. Simulation results showed that the estimated solute concentration values are very close to the achieved experimental results. The optimal conditions for emulsion liquid membrane were found to be: solvent kerosene, TOPO concentration (1%w/w), Amberlite LA-2 concentration (4%w/w), surfactant concentration (5%w/w), Na2CO3 concentration (5%w/v), modifier (decanol) concentration (2%w/w), mixing speed 300 rpm. The average accuracy percentages achieved by artificial neural network and k-nearest neighbor approaches were 88.75and#177;1.94% and 90.2and#177;1.2%, respectively.

Prediction of Succinic Acid Extraction Efficiency in the Emulsion Liquid Membrane by using Machine Learning Techniques

This research aims to predict succinic acid concentration in the external phase during the emulsion liquid membrane process by using artificial neural networks along with a popular alternative method: k-nearest neighbor technique. The solute concentration values can be predicted by the proposed method without performing a great number of emulsion liquid membrane experiments. Several computer simulations were performed to demonstrate the success of the system. Simulation results showed that the estimated solute concentration values are very close to the achieved experimental results. The optimal conditions for emulsion liquid membrane were found to be: solvent kerosene, TOPO concentration (1%w/w), Amberlite LA-2 concentration (4%w/w), surfactant concentration (5%w/w), Na2CO3 concentration (5%w/v), modifier (decanol) concentration (2%w/w), mixing speed 300 rpm. The average accuracy percentages achieved by artificial neural network and k-nearest neighbor approaches were 88.75and#177;1.94% and 90.2and#177;1.2%, respectively.

A Comparative Evaluation of Ensemble Classifiers for Malicious Webpage Detection

Malicious webpage is developed or manipulated to be used as attack tool where it is considered as one of the main reasons of Internet criminal activities. Thus, it is essential to detect such webpages and prevent end users form accessing it. The conventional malicious webpages detection techniques are based on searching through a blacklist that contains a list of webpages classified as malicious from the perspective of users. However, these techniques have high false-negative rates especially with aforesaid sophisticated attacks due to technical and computational limitations. Hence, machine learning techniques have been employed to classify webpages by systemically analyzing set of features that reflect the characteristics of a malicious webpage. This paper compares the prediction accuracy of several machine learning classification algorithms and ensemble techniques. A data set of 5000 instances of URLs, with 189 different features are used in the comparative study. The results show that the most accurate classification technique in MultiBoost and Adaboost is Support Vector Machine (SVM), while K-Nearest Neighbor (k-NN) technique in bagging and random subspace.

Wavelength Selection Method Based on Partial Least Square from Hyperspectral Unmanned Aerial Vehicle Orthomosaic of Irrigated Olive Orchards

Identifying and mapping irrigated areas is essential for a variety of applications such as agricultural planning and water resource management. Irrigated plots are mainly identified using supervised classification of multispectral images from satellite or manned aerial platforms. Recently, hyperspectral sensors on-board Unmanned Aerial Vehicles (UAV) have proven to be useful analytical tools in agriculture due to their high spectral resolution. However, few efforts have been made to identify which wavelengths could be applied to provide relevant information in specific scenarios. In this study, hyperspectral reflectance data from UAV were used to compare the performance of several wavelength selection methods based on Partial Least Square (PLS) regression with the purpose of discriminating two systems of irrigation commonly used in olive orchards. The tested PLS methods include filter methods (Loading Weights, Regression Coefficient and Variable Importance in Projection); Wrapper methods (Genetic Algorithm-PLS, Uninformative Variable Elimination-PLS, Backward Variable Elimination-PLS, Sub-window Permutation Analysis-PLS, Iterative Predictive Weighting-PLS, Regularized Elimination Procedure-PLS, Backward Interval-PLS, Forward Interval-PLS and Competitive Adaptive Reweighted Sampling-PLS); and an Embedded method (Sparse-PLS). In addition, two non-PLS based methods, Lasso and Boruta, were also used. Linear Discriminant Analysis and nonlinear K-Nearest Neighbors techniques were established for identification and assessment. The results indicate that wavelength selection methods, commonly used in other disciplines, provide utility in remote sensing for agronomical purposes, the identification of irrigation techniques being one such example. In addition to the aforementioned, these PLS and non-PLS based methods can play an important role in multivariate analysis, which can be used for subsequent model analysis. Of all the methods evaluated, Genetic Algorithm-PLS and Boruta eliminated nearly 90% of the original spectral wavelengths acquired from a hyperspectral sensor onboard a UAV while increasing the identification accuracy of the classification.

A semantic similarity based methodology for predicting protein-protein interactions: Evaluation with P53-interacting kinases.

Biomedical literature contains unstructured, rich information regarding proteins, ligands, diseases as well as biological pathways in which they are involved. Systematically analyzing such textual corpus has the potential for biomedical discovery of new protein-protein interactions and hidden drug indications. For this purpose, we have investigated a methodology that is based on a well-established text mining tool, Word2Vec, for the analysis of PubMed full text articles to derive word embeddings, and the use of a simple semantic similarity comparison either by itself or in conjunction with k-Nearest Neighbor (kNN) technique for the prediction of new relationships. To test this methodology, three lines of retrospective analyses of a dataset with known P53-interacting proteins have been conducted. First, we demonstrated that Word2Vec semantic similarity can infer functional relatedness among all kinases known to interact with P53. Second, in a series of time-split experiments, we demonstrated that both a simple similarity comparison and kNN models built with papers published up to a certain year were able to discover P53 interactors described in later publications. Third, in a different scenario of time-split experiments, we examined the predictions of P53-interacting proteins based on the kNN models built on data prior to a certain split year for different time ranges past that year, and found that the cumulative number of correct predictions was indeed increasing with time. We conclude that text mining of research papers in the PubMed literature based on Word2Vec analysis followed by a simple similarity comparison or kNN modeling affords excellent predictions of protein-protein interactions between P53 and kinases, and should have wide applications in translational biomedical studies such as repurposing of existing drugs, drug-drug interaction, and elucidation of mechanisms of action for drugs.

Reference-based model using multimodal gated recurrent units for image captioning

Describing images through natural language is a challenging task in the field of computer vision. Image captioning consists of creating image descriptions that can be accomplished via deep learning architectures that use convolutional neural networks (CNNs) and recurrent neural networks (RNNs). However, traditional RNNs encounter problems such as exploding and vanishing gradients, and they exhibit poor performance when generating non-descriptive sentences. To solve these issues, we proposed a model based on the encoder–decoder structure using CNNs to extract the image features and multimodal gated recurrent units (GRU) for descriptions. This model implements the part-of-speech (PoS) and likelihood function for weight generation in the GRU. The method performs knowledge transfer during a validation phase that uses the k-nearest neighbors technique (kNN). Experimental results using the Flickr30k and MSCOCO datasets demonstrated that the proposed PoS-based model presents competitive scores in comparison to state-of-the-art models. The system predicts more descriptive captions and closely approximates the expected captions both in the predicted and kNN selected captions.

Predicting adulteration of Palm oil with Sudan IV dye using shortwave handheld spectroscopy and comparative analysis of models

Shortwave handheld NIR spectroscopy coupled with multivariate algorithm was attempted to simultaneously classify and measure Sudan IV dye adulteration in palm oil samples. K-nearest neighbour (KNN) was used to develop a classification model to discriminate between authentic palm oil samples and Sudan IV dye adulterated (0.10−0.002 % w/w) ones. Principal component regression (PCR), partial least square regression (PLSR) and support vector machine regression (SVMR) algorithms were comparatively employed to quantify the addition of Sudan IV dye in authentic palm oil samples. The models were evaluated by the classification rate (R), correlation co-efficient in the calibration set/prediction set (Rp2/Rc2), and root mean square error of calibration/prediction (RMSEC/P). The developed multiplicative scatter correction plus KNN technique was found to accurately classify where R = 95.48 % and 97.00 % in the calibration set and prediction set respectively. Among the quantification model developed for measuring Sudan IV dye, standard normal variant preprocessing plus partial least square regression (SNV-PLSR) gave the best performance at Rc2 = 0.91, Rp2 = 0.90 and RMSEC = 0.0841, RMSEP = 0.0868 while standard normal variant preprocessing plus principal component regression gave Rc2 = 0.90, Rp2 = 0.90 and RMSEC = 0.0846, RMSEP = 0.870. The findings have proved that, the integrity of palm oil samples can be certified rapidly and nondestructively in terms of the presence of Sudan IV by using short wave handheld NIR spectroscopy. This offers the opportunity for incorporating NIR spectroscopy into mobile phone devices to enhance mobile detection.

Development of Chemical Oxygen on Demand (COD) Soft Sensor Using Edge Intelligence

The Internet of Things (IoT) based water quality monitoring system, mostly uses cost-effective sensors with a faster response time. Few water quality parameters such as Chemical Oxygen Demand (COD), Biological Oxygen Demand (BOD), Chlorine (Cl), and Total Phosphorous, hard to be measured online, soft sensing techniques are used as an alternate solution to this. This paper studies different Machine Learning (ML) models to prefer a suitable one that can identify the non-linearity of the dependency and establish a correlation between input and output parameters to design a COD soft sensor. The selected models are deployed in the proposed IoT architecture to predict the COD in real-time. More than 16000 data samples from the river Ganga with ten independent water quality parameters are collected and tested to validate the proposed IoT architecture. The paper compares the results of a few best performing ML algorithms like Multiple Linear Regression, Multilayer Perceptron, Support Vector Machine, Random Forest and K-Nearest Neighbour (KNN). Where, KNN technique proves to be the most efficient one in the prediction of COD in terms of response time and other performance matrices like R, R2, MSE, MAE, and RMSE. Finally, the preferred model (KNN) with the IoT setup is deployed and tested at the Sewage Treatment Plant (STP) outlet of the authors’ institute to verify the accuracy of the COD in real-time.

Lightweight Misbehavior Detection Management of Embedded IoT Devices in Medical Cyber Physical Systems

We propose a lightweight specification-based misbehavior detection management technique to efficiently and effectively detect misbehavior of an IoT device embedded in a medical cyber physical system through automatic model checking and formal verification. We verify our specification-based misbehavior detection technique with a patient-controlled analgesia (PCA) device embedded in a medical health monitoring system. Through extensive ns3 simulation, we verify its superior performance over popular machine learning anomaly detection methods based on support vector machine (SVM) and k-nearest neighbors (KNN) techniques in both effectiveness and efficiency performance metrics.

Feature Engineering for Predicting MOOC Performance

Increasing data recorded in massive open online course (MOOC) requires more automated analysis. The analysis, which includes making student’s prediction requires better strategy to produce good features and reduces prediction error. This paper presents the process of feature engineering for predicting MOOC student’s performance utilizing deep feature synthesis (DFS) method. The experiment produces features which all the top features selected using principal component analysis (PCA) are the features that are generated from method. In terms of prediction comparing both based features and generated features, the result shows better accuracy for generated features proposed using k-nearest neighbours technique which shows the method potential to be used for future prediction model.

Multi-Feature Approach for Bug Severity Assignment

When bug reports are submitted through bug tracking systems, they are analysed manually to identify their severity levels. A severity level specifies the negative impact of a bug on a system. With the huge number of submitted reports, setting the severity class manually is tedious and time consuming. Moreover, some bug types are reported more often than other types, which leads to imbalanced bug repositories. This paper proposes a multi-feature approach for automatic severity assignment, which leverages lexical, semantic, and categorical properties of the bug reports. The proposed approach utilizes word embeddings, topic model, vector space model, and an adapted K-Nearest Neighbour technique. Moreover, the impact of utilizing two sampling techniques, namely SMOTE and cluster-based under-sampling (CBU), were investigated. Experiments over two open source repositories, Eclipse and Mozilla, demonstrated that the proposed approach is superior to two previous studies.

RETRACTED: Efficient fuzzy based K-nearest neighbour technique for web services classification

Web services playing a vital role in the World Wide Web and generates huge amount of information across various domains of internet. Due to this evolution data in the form of articles, reports, digital galleries and web data of companies were increased everyday. To handle the huge volume of data each and every day, automatic query classification based on internet is more significant method. Research and development community has developed various techniques for the web services discovery, where it offers the mandated data for the improvement method. With respect to the literature survey, most of the researchers are concentrating to provide the efficient web service discovery. The amount of data that is available in the web is keeps on increasing and also it is used to differentiate the services, explanation and work of art. In order to achieve this method, machine learning algorithm is applied extensively for domain categorization. Various machine learning algorithm like KNN is applied for web service discovery. The systems are effectively learning the input and evaluate the performance accuracy with the given datasets. This paper, proposes an improved fuzzy with KNN algorithm for effective web service classification. This is used to increase an outcome in the form of accuracy and performance measures.

An Electric Eye for Human Activity Recognition: A Hybrid Neural Network

A real time detection of human movements is a practical solution to monitor aged people or mentally challenged people with the permission of their family. Household person is needed to monitor the elder and differently abled people. Instead of monitoring their activities with the help of other people, smart phones are used as a remote to monitor their activities and simultaneously send the message to their family members. The accelerometer sensor placed in the mobile phones. It is used to identify the activities of the person who holds the mobile phones. The most commonly used classifier technique is Naive Bayes classifier which has a limitation of handle with the large set of data. To overcome this defect, the proposed system classifies the data using k-nearest neighbor (K-NN) technique and Neuroevolution. This system recognize some representative human movements such as walking, climbing upstairs, climbing downstairs, standing, sitting and running ,using a conventional mobile equipped with a single tri-axial accelerometer sensor.