Advanced searches left 3/3

Categorical Data - Crossref

Summarized by Plex Scholar
Last Updated: 10 November 2022

* If you want to update the article please login/register

Modeling categorical time-to-event data: The example of social interaction dynamics captured with event-contingent experience sampling methods

These results enable the investigation of new types of scientific studies into the timing of those events, such as whether peopleu2019 affective state is related to the rate of social interactions and what types of social interactions are likely to take place. Although survival analysis methods have been used to analyze time-to-event data in longitudinal settings for many decades, new ones have yet to be integrated into ambulatory assessment studies. This paper explains how multilevel and multistate survival analysis techniques can be used to model the social interaction dynamics captured in large longitudinal studies, particularly when individuals exhibit particular types of behavior. We're giving you an introduction to these models and a guide on how to determine the timing and type of social interactions can be modeled using R statistical programming languages. This essay in sum outlines how survival models can help increase the recognition of dynamics that occur in everyday life.

Source link: https://doi.org/10.31234/osf.io/2hwvf


A fair-multicluster approach to clustering of categorical data

The fascination in designing fair clustering algorithms has risen. The main aim is to ensure that the output of a cluster algorithm is not biased toward or against specific subgroups of the population. We show that the Multicluster algorithm for clustering categorical data developed by Santos and Heras can be modified in order to improve the clusters' fairness. Of course, fairness and effectiveness must trade off, so an increase in the fairness goal often results in a loss of classification accuracy.

Source link: https://doi.org/10.1007/s10100-022-00824-2


Examining the Factors Affecting the Problem of Experiencing Difficulties while Online Shopping in Turkey with Categorical Data Analysis

Although there are numerous ways to express money, commodities, or services between the buyer and seller can be described as any step in the mutual exchange of money, commodities, or services between the buyer and seller through the use of internet technologies, online shopping can be described as any step in the reciprocal exchange of money, products, or services between the buyer and seller. The aim of this survey is to investigate the socioeconomic and economic factors that are relevant in cases where people have problems with their online purchases. The report used micro data from the 2021 Information and Communication Technology Usage Survey in Households, which was conducted by the Turkish Statistical Institute. Using binary logistic regression analysis, the contributing factors that led to the online shopping issues that individuals have with online purchases were determined. 56% of the men in the survey said they had issues with orders made through the website or mobile app in the last three months. 51. 9 percent of people in the eastern region reported having trouble with payments made through the website or mobile app, according to the report.

Source link: https://doi.org/10.54709/iisbf.1152952


Process control for categorical (ordinal) data

Quality improvement is playing a vital role in the growth of a company. Hence, developing control charts for tracking ordinal data has become a recent research area. Since there are quite a few techniques available in the literature for this purpose, quality control practitioners often have a challenge trying to determine the appropriate method for monitoring ordinal data in the practical field. For various values of shift in mean using different methodologies under investigation, simulations were simulated from Normal distribution and average run length were estimated for various amounts of change in mean.

Source link: https://doi.org/10.4314/ijest.v14i2.4


Embedding-Based Complex Feature Value Coupling Learning for Detecting Outliers in Non-IID Categorical Data

In realworld applications, Non-IID categorical data is ubiquitous and common. Existing outlier detection schemes typically only focus on pairwise primary value couplings, failing to reveal real connections that hide in complicated couplings, resulting in suboptimal and unstable results. An embedding scheme is implemented on the value network built via biased value couplings, which aids in the understanding of high-order complex value couplings and embeds them into a value representation matrix. It is suggested that bidirectional selective value coupling learning be used to show how to quantify value and object outlierness by value couplings.

Source link: https://doi.org/10.1609/aaai.v33i01.33015541


Categorical edge-based analyses of phylogenomic data reveal conflicting signals for difficult relationships in the avian tree

Phylogenetic studies fail to produce a satisfactory conclusion of some tree of life relationships even with genome-scale databases, so the failure is unlikely to reflect data quality limitations. Here, we conduct an edge-based study of Neoavian evolution, examining the phylogenetic validity of two new phylogenomic bird datasets and three datatypes. On each of the nodes we examined, Edge-based studies increased congruence and revealed more detail about the effects of data type, GC content variation, and outlier genes. First, outlier gene genes appeared to have different patterns of support for the relationships among the earliest diverging Neoaves. Hoatzin's placement was highly variable, but our EBA did report a previously unappreciated data type effect with a huge effect on its rank. The owls + mousebirds heard at a distance of 1000 miles from a high GC CV loci, which boosted the call for owls + mousebirds.

Source link: https://doi.org/10.1101/2021.05.17.444565


Diving Deeper Into Categorical Data

Using the Common Online Data Analysis Platform, we designed an afterschool program in which students explored existing data. We'll discuss how students participated in exploratory data analysis that involved comparing two or more categorical variables of concern, as well as complicated comparisons in a large contingency table. This report is one of the first to examine categorical data that are not pre-structured.

Source link: https://doi.org/10.52041/iase.icots11.t2i1


Categorical Data Clustering Using Harmony Search Algorithm for Healthcare Datasets

Many benefits in healthcare dashboard software exist, including healthcare statistics. Harmony search based categorical clustering was suggested in this paper for an improved clustering for healthcare data. Generally, researchers use genetic algorithm clustering algorithms to find locally optimal solutions to global optimal solutions. The HSCC's dental and 71% lung cancer survey shows 98% accuracy for dental and 71%. Although GACC has a 95% and 65% success for dental data and lung cancer studies, the GACC has a 95% and 65% success.

Source link: https://doi.org/10.4018/ijehmc.309440


Clustering heterogeneous categorical data using enhanced mini batch K-means with entropy distance measure

Clustering techniques in data mining aims to identify a set of patterns based on their similarity. With various scales of data scales including nominal, ordinal, binary, and Likert scales, heterogeneous results are collected in a data analysis. For the heterogeneous categorical results, the latest entropy distance survey seems to have good results. This paper introduces a hypothesis for heterogeneous categorical data solutions based on a mini batch k-means with entropy compensation, which is intended to determine the success of similarity testing in clustering method using heterogeneous categorical data. Fowlkes-Mallows' index stood at 0. 94, better than other clustering algorithms with the precision at 0. 88, adjusted rand index at 0. 87, and Fowlkes-Mallow index at 0. 94. Cluster generation, k at 0. 26 s, is the highest elapsed time for cluster generation in several domains, according to researchers. The new strategy may be useful for increasing the quality of clustering for heterogeneous categorical data problems in several domains.

Source link: https://doi.org/10.11591/ijece.v13i1.pp1048-1059

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions

* Please keep in mind that all text is summarized by machine, we do not bear any responsibility, and you should always check original source before taking any actions