scholarly journals Towards Improving Predictive Statistical Learning Model Accuracy by Enhancing Learning Technique

2022 ◽  
Vol 42 (1) ◽  
pp. 303-318
Author(s):  
Ali Algarni ◽  
Mahmoud Ragab ◽  
Wardah Alamri ◽  
Samih M. Mostafa
Keyword(s):  
Statistical Learning ◽  
Learning Model ◽  
Model Accuracy ◽  
Learning Technique

scholarly journals Assessing the frontier: Active learning, model accuracy, and multi-objective candidate discovery and optimization

2020 ◽  
Vol 153 (2) ◽  
pp. 024112 ◽  
Author(s):  
Zachary del Rosario ◽  
Matthias Rupp ◽  
Yoolhee Kim ◽  
Erin Antono ◽  
Julia Ling
Keyword(s):  
Active Learning ◽  
Learning Model ◽  
Model Accuracy ◽  
Multi Objective

scholarly journals Pengenalan Gerak Manusia Menggunakan Algoritma Relevance Vector Machine pada MSRC-12 Dataset

2020 ◽  
Vol 3 (1) ◽  
Author(s):  
Vina Ayumi ◽  
Erwin Dwika Putra
Keyword(s):  
Machine Learning ◽  
Data Processing ◽  
Statistical Learning ◽  
Gesture Recognition ◽  
Learning Theory ◽  
Relevance Vector Machine ◽  
Motion Model ◽  
Communication Tools ◽  
Machine Learning Technique ◽  
Learning Technique

Relevance vector machine is a popular machine learning technique that is motivated by statistical learning theory. RVM can be used for gesture recognition which is one of the communication tools used by humans. This study proposes an experiment using the Relevance Vector Machine (RVM) algorithm on gesture data from Microsoft Research Cambridge-12 (MSRC-12) as a proposed solution to overcome unbalanced problems in data processing. The results of the study are the accuracy for 1-person motion model reaches 100% and the lowest accuracy with 5 people the motion model reaches 96%. Graphically, the more people or models, the lower the algorithm's accuracy.

scholarly journals Simulating language-specific and language-general effects in a statistical learning model of Chinese reading

2009 ◽  
Vol 61 (2) ◽  
pp. 238-257 ◽  
Author(s):  
Jianfeng Yang ◽  
Bruce D. McCandliss ◽  
Hua Shu ◽  
Jason D. Zevin
Keyword(s):  
Statistical Learning ◽  
Learning Model ◽  
Chinese Reading

Shallow learning model for diagnosing neuro muscular disorder from splicing variants

2017 ◽  
Vol 14 (4) ◽  
pp. 329-336
Author(s):  
Sathyavikasini Kalimuthu ◽  
Vijaya Vijayakumar
Keyword(s):  
Pattern Recognition ◽  
Muscular Dystrophy ◽  
Statistical Learning ◽  
Positional Cloning ◽  
K Nearest Neighbor ◽  
Gene Sequences ◽  
Content Type ◽  
Splicing Mutations ◽  
Learning Technique ◽  
Cloned Gene

Purpose Diagnosing genetic neuromuscular disorder such as muscular dystrophy is complicated when the imperfection occurs while splicing. This paper aims in predicting the type of muscular dystrophy from the gene sequences by extracting the well-defined descriptors related to splicing mutations. An automatic model is built to classify the disease through pattern recognition techniques coded in python using scikit-learn framework. Design/methodology/approach In this paper, the cloned gene sequences are synthesized based on the mutation position and its location on the chromosome by using the positional cloning approach. For instance, in the human gene mutational database (HGMD), the mutational information for splicing mutation is specified as IVS1-5 T > G indicates (IVS - intervening sequence or introns), first intron and five nucleotides before the consensus intron site AG, where the variant occurs in nucleotide G altered to T. IVS (+ve) denotes forward strand 3′– positive numbers from G of donor site invariant and IVS (−ve) denotes backward strand 5′ – negative numbers starting from G of acceptor site. The key idea in this paper is to spot out discriminative descriptors from diseased gene sequences based on splicing variants and to provide an effective machine learning solution for predicting the type of muscular dystrophy disease with the splicing mutations. Multi-class classification is worked out through data modeling of gene sequences. The synthetic mutational gene sequences are created, as the diseased gene sequences are not readily obtainable for this intricate disease. Positional cloning approach supports in generating disease gene sequences based on mutational information acquired from HGMD. SNP-, gene- and exon-based discriminative features are identified and used to train the model. An eminent muscular dystrophy disease prediction model is built using supervised learning techniques in scikit-learn environment. The data frame is built with the extracted features as numpy array. The data are normalized by transforming the feature values into the range between 0 and 1 aid in scaling the input attributes for a model. Naïve Bayes, decision tree, K-nearest neighbor and SVM learned models are developed using python library framework in scikit-learn. Findings To the best knowledge of authors, this is the foremost pattern recognition model, to classify muscular dystrophy disease pertaining to splicing mutations. Certain essential SNP-, gene- and exon-based descriptors related to splicing mutations are proposed and extracted from the cloned gene sequences. An eminent model is built using statistical learning technique through scikit-learn in the anaconda framework. This paper also deliberates the results of statistical learning carried out with the same set of gene sequences with synonymous and non-synonymous mutational descriptors. Research limitations/implications The data frame is built with the Numpy array. Normalizing the data by transforming the feature values into the range between 0 and 1 aid in scaling the input attributes for a model. Naïve Bayes, decision tree, K-nearest neighbor and SVM learned models are developed using python library framework in scikit-learn. While learning the SVM model, the cost, gamma and kernel parameters are tuned to attain good results. Scoring parameters of the classifiers are evaluated using tenfold cross-validation using metric functions of scikit-learn library. Results of the disease identification model based on non-synonymous, synonymous and splicing mutations were analyzed. Practical implications Certain essential SNP-, gene- and exon-based descriptors related to splicing mutations are proposed and extracted from the cloned gene sequences. An eminent model is built using statistical learning technique through scikit-learn in the anaconda framework. The performance of the classifiers are increased by using different estimators from the scikit-learn library. Several types of mutations such as missense, non-sense and silent mutations are also considered to build models through statistical learning technique and their results are analyzed. Originality/value To the best knowledge of authors, this is the foremost pattern recognition model, to classify muscular dystrophy disease pertaining to splicing mutations.

scholarly journals Developing a Flipped Learning model for Teaching EAP Students Struggling with Multi-Level Challenges in a College Context

2020 ◽  
Vol 9 (6) ◽  
pp. 1
Author(s):  
Iman Alkhalidi
Keyword(s):  
Blended Learning ◽  
Classroom Practice ◽  
Learning Model ◽  
English For Academic Purposes ◽  
Learning Theories ◽  
Flipped Learning ◽  
Holistic Model ◽  
Online Engagement ◽  
Learning Technique ◽  
Multi Level

This article aims to understand students’ experiences regarding the implementation of flipped learning (FL) as a modern blended learning technique in teaching English for academic purposes (EAP) in a community college context in Toronto. Based on students’ views, blended learning theories, and several previous studies, the study also aims to develop a holistic contextualized flipped learning model that helps both students and teachers in the context of EAP to cope with the challenges of a multilevel EAP classroom. The study is guided by the epistemology and philosophy of the interpretive paradigm as an underpinning stance. Accordingly, the qualitative approach has been selected for determining the strategy and methods of sampling, and data collection and data analysis. Results revealed that students’ views are compatible with the theoretical views in validating the utilization of flipped learning as a modern technique in the context of EAP. However, results revealed that the development of a holistic model includes a further component-online engagement as an extension component to the model. The study offers a set of recommendations and implications for EAP teachers and instructors within the area of ELT for classroom practice.

scholarly journals PEMBELAJARAN UNTUK MENGEMBANGKAN KEMAMPUAN PENALARAN INFERENSIAL STATISTIS

2018 ◽  
Vol 6 (1) ◽  
pp. 16
Author(s):  
Laila Hayati
Keyword(s):  
Learning Environment ◽  
Statistical Learning ◽  
Learning Theory ◽  
Thinking Skills ◽  
Learning Model ◽  
Statistical Reasoning ◽  
Inferential Reasoning ◽  
Reasoning Abilities ◽  
Informal Inferential Reasoning ◽  
Definition Of

Abstrak: Artikel ini membahas tentang pembelajaran yang dapat digunakan untuk mengembangkan kemampuan penalaran inferensial statistis. Tujuan kognitif pembelajaran Statistika adalah untuk mengembangkan kemampuan literasi, penalaran, dan berpikir statistis. Salah satu tujuan dalam penalaran statistis adalah mengembangkan kemampuan penalaran inferensial. Model pembelajaran yang dapat digunakan untuk mengembangkan kemampuan penalaran inferensial statistis adalah model  Statistical Reasoning Learning Environment  (SRLE) yang didasarkan pada teori belajar konstruktivisme. Analisis didasarkan pada: 1) definisi inferensi statistis; 2) definisi penalaran inferensial informal dan formal; 3) kerangka kerja penalaran inferensial informal; 5) model pembelajaran SRLE; dan 6) penelitian terkait penalaran inferensial statistis.Abstract: This article deals with learning that can be used to develop inferential reasoning abilities statistically. The cognitive goal of statistical learning is to develop literacy, reasoning, and statistical thinking skills. One of the goals in statistical reasoning is to develop inferential reasoning abilities. The learning model that can be used to develop static inferential reasoning abilities is Statistical Reasoning Learning Environment (SRLE) model based on constructivism learning theory. The analysis is based on: 1) the definition of statistical inference; 2) the definition of informal and formal inferential reasoning; 3) informal inferential reasoning framework; 5) the learning model of SRLE; and 6) research related to inferential statistical reasoning.

scholarly journals Beyond Traditional Measures of Personality with BISCUIT and BARE: A New Statistical Learning Technique and Behavioral Item Pool to Push Personality Psychology Forward

2020 ◽  
Author(s):  
Lorien Grey Elleman
Keyword(s):  
Statistical Learning ◽  
Big Five ◽  
Traditional Approach ◽  
Item Type ◽  
Item Pool ◽  
Pilot Studies ◽  
Personality Psychology ◽  
Learning Technique ◽  
Statistical Learning Techniques ◽  
Behavioral Frequencies

This dissertation investigates two ways in which personality psychology should move beyond the traditional approach of measuring personality with broad domains composed of trait descriptors, as exemplified by the Big Five taxonomy. The first study (Chapter 2) suggests an alternative to the traditional approach of aggregating personality items into domains. Mounting evidence indicates that, compared to domains, narrower measures of personality account for more variance in criteria and describe personality-criterion relationships more accurately. Analysis of individual personality items is the most granular approach to studying personality and is typically performed with statistical learning techniques (SLTs). The first study: (a) champions a new statistical learning technique, BISCUIT; (b) finds that BISCUIT provides a balance between prediction and parsimony; and (c) replicates previous findings that the broadness of the Big Five traits hinder their predictive power.The second study (Chapter 3) suggests an alternative to the traditional approach of measuring personality with trait descriptors, or "traditional personality items." Of the three patterns commonly associated with personality (cognitions, emotions, and behaviors), behaviors are the least studied; traditional personality items tend to measure cognitions and emotions. Historically, yearlong patterns of specific behaviors have been thought of as criteria of personality measures, but the second study posits they should be classified as personality items because they measure patterns of behavior, a component of personality. The second study reviews and extends two pilot studies that indicated behavioral frequencies predict life outcomes, sometimes better than traditional personality items. The second study: (a) estimates the extent to which behavioral frequencies strengthen personality-criterion relationships above traditional personality items; (b) determines that some criteria are differentially predicted by personality item type; and (c) publishes an updated, public-domain item pool of behavioral frequencies: the BARE (Behavioral Acts, Revised and Expanded) Inventory.

scholarly journals Effective Feature Selection Strategy Using Manova Test

2019 ◽  
Vol 8 (2) ◽  
pp. 5969-5971
Keyword(s):  
Feature Selection ◽  
Learning Model ◽  
Statistical Test ◽  
Classification Model ◽  
Selection Strategy ◽  
Model Accuracy ◽  
Increasing Demand ◽  
New Feature ◽  
The Right ◽  
Day By Day

Feature selection is the most important step to develop any latest learning model. As the complexity of the leaning models increases day by day there is an increasing demand, in selecting the right features to build the model. There are many methods for feature selection. A new feature selection based on the Manova statistical test is implemented. Using the Manova test, we select attributes from academic datasets. Using the selected attributes, we build a classification model. Accuracy of the model with feature selection is compared with a model with all attributes. Results are discussed. It is proved that the classification model build with features selected by Manova test achieves more accuracy than a model built with all features.

Sign in / Sign up

Export Citation Format

Share Document