scholarly journals Pengenalan Bentuk Tangan secara Real Time Menggunakan Leap Motion dan K-Nearest Neighbors sebagai Sistem Kendali Robot Beroda

2020 ◽  
Vol 5 (2) ◽  
pp. 178
Author(s):  
Supria Supria ◽  
Depandi Enda ◽  
Muhamad Nasir
Keyword(s):  
Real Time ◽  
Euclidean Distance ◽  
Nearest Neighbors ◽  
Leap Motion ◽  
K Nearest Neighbors ◽  
Accelerometer Sensor ◽  
Finger Tip

Sistem kendali robot saat ini telah banyak dibuat dengan menggunakan berbagai metode seperti sensor accelerometer, sensor suara, leap motion. Pada penelitian ini diusulkan pengenalan bentuk tangan secara real time menggunakan leap motion dan K-Nearest Neighbors (KNN) sebagai sistem kendali robot beroda. Leap motion digunakan untuk mendeteksi titik koordinat posisi tangan pada pandangan leap motion. pembentukan fitur dilakukan dengan mengukur jarak euclidean distance antara palm position dengan finger tip. KNN digunakan untuk menentukan kelas pada data testing. Uji coba dilakukan menggunakan tangan penulis dengan 5 jenis instruksi yaitu maju, mundur, stop, belok kanan, belok kiri dan setiap instruksi di ujicoba sebanyak 20 kali percobaan. Dari hasil ujicoba yang dilakukan menunjukkan bahwa sistem yang diusulkan memiliki tingkat akurasi pengenalan 94%.

Klasifikasi Gerakan Jatuh Berbasis Accelerometer dan Gyroscope Menggunakan K-Nearest Neighbors

2020 ◽  
Vol 4 (2) ◽  
pp. 24-29
Author(s):  
Adlian Jefiza ◽  
Indra Daulay ◽  
Jhon Hericson Purba
Keyword(s):  
Real Time ◽  
Root Mean Square ◽  
Nearest Neighbors ◽  
Training Data ◽  
Mean Square ◽  
K Nearest Neighbors ◽  
Testing Data ◽  
Total Data

Permasalahan utama pada penelitian ini merujuk kepada semakin menurunnya daya tahan tubuh lanjut usia (lansia). Hal ini membutuhkan sistem monitoring aktivitas lansia secara real time. Untuk mendeteksi kegiatan para lansia, dirancang sebuah perangkat monitoring dengan accelerometer 3-sumbu dan gyroscope 3-sumbu. Data sensor diperoleh dari lima partisipan. Setiap partisipan melakukan lima gerakan yaitu terjatuh, duduk, tidur, rukuk dan sujud. Gerakan yang dipilih adalah gerakan yang menyerupai gerakan jatuh. Total data yang diperoleh dari partisipan adalah 75 data yang terbagi menjadi training data dan testing data. Penelitian ini menggunakan metode transformasi Wavelet untuk mengenali fitur dari gerakan. Untuk pengklasifikasian setiap gerakan, digunakan metode K-nearest neighbors (KNN). Hasil klasifikasi gerakan menggunakan lima kelas menghasilkan nilai root mean square sebesar 0.0074 dengan akurasi 100%.

Personality Based Lipstick Color Recommender System using K-Nearest Neighbors Algorithm

2019 ◽  
Vol 6 (1) ◽  
pp. 55-59
Author(s):  
Ryan Adiputra ◽  
Ni Made Satvika Iswari ◽  
Wella Wella
Keyword(s):  
Recommender System ◽  
Euclidean Distance ◽  
Nearest Neighbors ◽  
Training Data ◽  
Personality Test ◽  
Myers Briggs Type Indicator ◽  
K Nearest Neighbors ◽  
Type Indicator ◽  
Myers Briggs ◽  
The Right

Lipstick is a lip color which available in many colors. A research said instant valuation of woman personality can be figured by their lipstick color choice. Therefore there is a necessity to use the right lipstick color to obtain a harmony between personality and appearance. This experiment was conducted to give lipstick color recommendation by using K-Nearest Neighbors algorithm, and Myers-Briggs Type Indicator (MBTI) personality test instrument. The system was built on Android application. Euclidean distance value is affected by 5 factors which are age, introvert, sensing, thinking, and judging. Lipstick color recommendation is obtained by fetching 7 training data with nearest Euclidean distance when compared to personality test result. The colors used in this experiment are nude, pink, red, orange, and purple. After evaluation, it is obtained the application’s accuracy of 87.38% which considered as good classification, both precision and recall with 75.68% which considered as fair classification. The score for software quality is 79.13% which considered as good quality. Keywords—K-Nearest Neighbors, Data Mining, Myers-Briggs Type Indicator,Recommender System, Lipstick.  

scholarly journals Accelerating the K-Nearest Neighbors Filtering Algorithm to Optimize the Real-Time Classification of Human Brain Tumor in Hyperspectral Images

Sensors ◽  
2018 ◽  
Vol 18 (7) ◽  
pp. 2314 ◽  
Author(s):  
Giordana Florimbi ◽  
Himar Fabelo ◽  
Emanuele Torti ◽  
Raquel Lazcano ◽  
Daniel Madroñal ◽  
...  
Keyword(s):  
Real Time ◽  
Nearest Neighbors ◽  
K Nearest Neighbors ◽  
Human Brain Tumor ◽  
Real Time Processing ◽  
Surgical Guidance ◽  
Filtering Algorithm ◽  
Sequential Implementation ◽  
Very High

The use of hyperspectral imaging (HSI) in the medical field is an emerging approach to assist physicians in diagnostic or surgical guidance tasks. However, HSI data processing involves very high computational requirements due to the huge amount of information captured by the sensors. One of the stages with higher computational load is the K-Nearest Neighbors (KNN) filtering algorithm. The main goal of this study is to optimize and parallelize the KNN algorithm by exploiting the GPU technology to obtain real-time processing during brain cancer surgical procedures. This parallel version of the KNN performs the neighbor filtering of a classification map (obtained from a supervised classifier), evaluating the different classes simultaneously. The undertaken optimizations and the computational capabilities of the GPU device throw a speedup up to 66.18× when compared to a sequential implementation.

Towards Real-Time Illegal Logging Monitoring: Gas-Powered Chainsaw Logging Detection System using K-Nearest Neighbors

2020 ◽  
Author(s):  
John Daniel C. Arevalo ◽  
Pauline C. Calica ◽  
Bernadette Andree D. R. Celestino ◽  
Katami A. Dimapunong ◽  
Dylan Josh D. Lopez ◽  
...  
Keyword(s):  
Real Time ◽  
Detection System ◽  
Nearest Neighbors ◽  
Illegal Logging ◽  
K Nearest Neighbors

scholarly journals Klasifikasi Sekolah Menengah Pertama/Sederajat Wilayah Bireuen Menggunakan Algoritma K-Nearest Neighbors Berbasis Web

2020 ◽  
Vol 5 (1) ◽  
pp. 33
Author(s):  
Rozzi Kesuma Dinata ◽  
Fajriana Fajriana ◽  
Zulfa Zulfa ◽  
Novia Hasdyna
Keyword(s):  
Euclidean Distance ◽  
Nearest Neighbor ◽  
Nearest Neighbors ◽  
K Nearest Neighbor ◽  
K Nearest Neighbors

Pada penelitian ini diimplementasikan algoritma K-Nearest Neighbor dalam pengklasifikasian Sekolah Menengah Pertama/Sederajat berdasarkan peminatan calon siswa. Tujuan penelitian ini adalah untuk memudahkan pengguna dalam menemukan sekolah SMP/sederajat berdasarkan 8 kriteria sekolah yaitu akreditasi, fasilitas ruangan, fasilitas olah raga, laboratorium, ekstrakulikuler, biaya, tingkatan kelas dan waktu belajar. Adapun data yang digunakan dalam penelitian ini didapatkan dari Dinas Pendidikan Pemuda dan Olahraga Kabupaten Bireuen. Hasil penelitian dengan menggunakan K-NN dan pendekatan Euclidean Distance dengan k=3, diperoleh nilai precision sebesar 63,67%, recall 68,95% dan accuracy sebesar 79,33% .

scholarly journals Real Time Efficient Accident Predictor System using Machine Learning Techniques (kNN, RF, LR, DT)

2020 ◽  
Vol 10 (2) ◽  
pp. 108-111
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Real Time ◽  
Classification Accuracy ◽  
Nearest Neighbors ◽  
Machine Learning Algorithms ◽  
Machine Learning Techniques ◽  
Classification Methods ◽  
K Nearest Neighbors ◽  
Learning Techniques

Real time crash predictor system is determining frequency of crashes and also severity of crashes. Nowadays machine learning based methods are used to predict the total number of crashes. In this project, prediction accuracy of machine learning algorithms like Decision tree (DT), K-nearest neighbors (KNN), Random forest (RF), Logistic Regression (LR) are evaluated. Performance analysis of these classification methods are evaluated in terms of accuracy. Dataset included for this project is obtained from 49 states of US and 27 states of India which contains 2.25 million US accident crash records and 1.16 million crash records respectively. Results prove that classification accuracy obtained from Random Forest (RF) is96% compared to other classification methods.

scholarly journals Incident forecasting model for motorcycle driving based on IoT and artificial intelligence

2021 ◽  
Vol 24 (1) ◽  
pp. 444
Author(s):  
Esteban Alejandro Cárdenas-Lancheros ◽  
Nelson Enrique Vera-Parra
Keyword(s):  
Artificial Intelligence ◽  
Nearest Neighbors ◽  
Forecasting Model ◽  
Accident Risk ◽  
Classification Models ◽  
Road Accidents ◽  
K Nearest Neighbors ◽  
Accelerometer Sensor ◽  
Technological Tools ◽  
And Storage

Internet of things (IoT) and artificial intelligence provide more and more solutions to the exercise of capturing data effectively, taking them through processing and analysis stages to extract valuable information. Currently, technological tools are applied to counteract incidents in motorcycle driving, whether they are part of the same vehicle or are externally involved in the environment. Incidents in motorcycle driving are increasing due to the demand for the acquisition of these vehicles, which makes it important to generate an approach towards reducing the risk of road accidents based on the analysis of dynamic behavior while driving. The development of this research began with the detection and storage of data associated with the dynamic acceleration variable of a motorcycle while driving, this with the help of a 3-axis accelerometer sensor generating a dataset, which was processed and analyzed for later be taken by three predictive classification models based on machine learning which were decision trees, K-Nearest neighbors and random forests. The performance of each model was evaluated in the task of better classifying the level of accident risk, concerning the driving style based on certain levels of acceleration. The random forest model showed a slightly better performance compared to that shown by the other two models, with 97.24% accuracy and recall, 97.16% precision and 97.17% F1 score.

scholarly journals Approximate k-nearest neighbors graph for single-cell Hi-C dimensional reduction with MinHash

2020 ◽  
Author(s):  
Joachim Wolff ◽  
Rolf Backofen ◽  
Björn Grüning
Keyword(s):  
Single Cell ◽  
Dimensional Reduction ◽  
Euclidean Distance ◽  
Nearest Neighbors ◽  
High Dimensional ◽  
K Nearest Neighbors ◽  
Data Set ◽  
Distance Computation ◽  
Approximate Nearest Neighbors ◽  
Interaction Matrices

Single-cell Hi-C interaction matrices are high dimensional and very sparse. To cluster thousands of single-cell Hi-C interaction matrices they are flattened and compiled into one matrix. This matrix can, depending on the resolution, have a few millions or even billions of features and any computation with it is therefore memory demanding. A common approach to reduce the number of features is to compute a nearest neighbors graph. However, the exact euclidean distance computation is in O(n2) and therefore we present an implementation of an approximate nearest neighbors method based on local sensitive hashing running in O(n). The presented method is able to process a 10kb single-cell Hi-C data set with 2500 cells and needs 53 GB of memory while the exact k-nearest neighbors approach is not computable with 1 TB of memory.

Sign in / Sign up

Export Citation Format

Share Document