Large high dimensional data handling using data reduction

Patel, Falguni N.

doi:10.1109/iceeot.2016.7754940

Cited by 6 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The vector representation of the words defines the position of each word in a high dimensional space, typically 100 to 500 dimensions. However, high-dimensional data is hard to visualize making it hard to identify what words similar to each other (Patel, 2016). Dimensionality reduction techniques reveal the underlying structure of the data.…”

Section: A Primer On Text Analysismentioning

confidence: 99%

Measuring trust: A text analysis approach to compare, contrast, and select trust questionnaires

Areen¹,

Li²,

Chiou³

et al. 2022

Preprint

View full text Add to dashboard Cite

Trust has emerged as a prevalent construct to describe relationships between people and between people and technology in myriad domains. Across disciplines, researchers have relied on many different questionnaires to measure trust. The degree to which these questionnaires differ has not been systematically explored. In this paper, we use a word-embedding text analysis technique to identify the differences and common themes across the most used trust questionnaires and provide guidelines for questionnaire selection. A mapping review was conducted to identify the existing trust questionnaires. In total, we included 46 trust questionnaires from three main domains (i.e., Automation, Humans, and E-commerce) with a total of 626 items measuring different trust layers (i.e., Dispositional, Learned, and Situational). Next, we encoded the words within each questionnaire using GloVe word embeddings and computed the embedding for each questionnaire item, and for each questionnaire. We reduced the dimensionality of the resulting dataset using UMAP to visualize these embeddings in scatterplots and implemented the visualization in a web app for interactive exploration of the questionnaires (link to app). At the word level, the semantic space serves to produce a lexicon of trust-related words. At the item and questionnaire level, the analysis provided recommendation on questionnaire selection based on the dispersion of questionnaires’ items and at the domain and layer composition of each questionnaire. Along with the web app, the results help explore the semantic space of trust questionnaires and guide the questionnaire selection process. The results provide a novel means to compare and select trust questionnaires and to glean insights about trust from spoken dialog or written comments.

show abstract

Section: A Primer On Text Analysismentioning

confidence: 99%

Measuring trust: A text analysis approach to compare, contrast, and select trust questionnaires

Areen¹,

Li²,

Chiou³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Чим менше часу витрачається на очікування, тим більш ефективна робота вебзастосунку [1][2][3][4]. У роботі [5] розглядаються такі методи, як використання файлів cookie, зберігання сеансів, локальне зберігання та indexDB. Автори обговорюють переваги й обмеження кожного із цих підходів та їх ефективність.…”

Section: вступunclassified

Software Methods Analysis of Optimizing the Local State of Data Web Applications Management

Оleshchenko¹

2022

KPI SN

View full text Add to dashboard Cite

Background. The use of various libraries often leads to a decrease in the speed of the web application and the complexity of the execution of the program code. The object of this study is the process of storing and managing the data of the client part of the web application, the subject of the research is the software methods of managing the local state of the data of the web application. Objective. The goal of the article is the reduction in the data processing time of web applications relative to existing software methods. Methods. The main idea of the proposed method is to use an atomic approach to the state of the web application data. Having an arbitrary entity, in the general state of the web application, a state fragment is created that is responsible only for this entity. Such a fragment is independent of other state fragments and can only work with the encapsulated entity. Using encapsulation, the configuration of an entity is passed to the React Context API as an object containing data and functions that modify it. Results. The developed framework method was compared with popular state management libraries Redux, MobXState-Tree and Recoil. Comparing each of the test scenarios in a percentage ratio, an average decrease in program execution time by 17% was obtained. Conclusions. The Redux, MobXState-Tree, and Recoil libraries were selected for the research of existing software methods. The analysis of methods was performed using the SonarQube utility. To evaluate the results of the software methods, the Google Chrome browser utility DevTools was used. The proposed optimized software method allows to reduce the data processing time and optimize the state management of web applications. Keywords: web application, optimization, local state of data, Redux, MobXState-Tree, Recoil.

show abstract

“…The weighted k-nearest neighbor classifier was utilized to prove the importance of data reduction. After using uniform random sampling selection for data reduction in both direction-instances and attributes, accuracy is preserved, and an increase of execution time was observed [22].…”

Section: Rank-based Variable Minimization Using Clustering Algorithmmentioning

confidence: 99%

Rank-Based Variable Minimization Using Clustering Algorithm

Delima¹,

Francisco²

2019

IJMLC

View full text Add to dashboard Cite

This paper introduces variable minimization with the advent of data mining techniques, particularly the cluster analysis using the K-means algorithm. The results of the simulation process validated the actual ranking of the variables in each category from the 407 records of student-respondents. A noticeable decrease of variables in determining the instructional performance of teachers in the Caraga Region, Philippines is evident in the simulation results. The variables that were clustered by the algorithm that conforms to have the lowest rank based on the actual survey was removed. Results showed that there were a total of fourteen (14) variables removed from the thirty-item survey questionnaire. In this case, it only proved that like any other data mining algorithms, the cluster analysis, particularly with the K-Means algorithm, is also effective in determining what variables in the data set to omit based on the groupings with the lowest rank it generated.

show abstract

Large high dimensional data handling using data reduction

Cited by 6 publications

References 13 publications

Measuring trust: A text analysis approach to compare, contrast, and select trust questionnaires

Measuring trust: A text analysis approach to compare, contrast, and select trust questionnaires

Software Methods Analysis of Optimizing the Local State of Data Web Applications Management

Rank-Based Variable Minimization Using Clustering Algorithm

Contact Info

Product

Resources

About