Digital Commons Network^™

A Medical Data Cleaner, Jahnavi Yetukuri

All Graduate Plan B and other Reports, Spring 1920 to Spring 2023

This report describes medical-data cleaning tool, called MedDataCleaner that can detect outliers in medical data and assistant Database Administrators in resolving data-related problem. Specifically, MedDataCleaner, enables the users to define cleaning rules and offers the ability to choose classification methods that help determine if the data is good or bad. MedDataClearer uses Vitruvian DB objects for object-relation mapping (ORM) support and Vitruvian alignment links for designing the GUI.

My contribution towards this work includes designing the user interfaces using Vitruvian Alignment links, design and implement mean, standard deviation and neural classification methods using Vitruvian DB objects.

Windowed Pq-Grams For Approximate Joins Of Data-Centric Xml, Nikolaus Augsten, Michael H. Böhlen, Curtis Dyreson, Johann Gamper

Curtis Dyreson

In data integration applications, a join matches elements that are common to two data sources. Since elements are represented slightly different in each source, an approximate join must be used to do the matching. For XML data, most existing approximate join strategies are based on some ordered tree matching technique, such as the tree edit distance. In data-centric XML, however, the sibling order is irrelevant, and two elements should match even if their subelement order varies. Thus, approximate joins for data-centric XML must leverage unordered tree matching techniques. This is computationally hard since the algorithms cannot rely on a predefined …

Optimal Candidate Generation In Spatial Co-Location Mining, Zhongshan Lin

All Graduate Theses and Dissertations, Spring 1920 to Summer 2023

Existing spatial co-location algorithms based on levels suffer from generating extra, nonclique candidate instances. Thus, they require cliqueness checking at every level. In this thesis, a novel, spatial co-location mining algorithm that automatically generates co-located spatial features without generating any nonclique candidates at any level is proposed. Subsequently, this algorithm generates fewer candidates than other existing level-wise, co-location algorithms without losing any pertinent information. The benefits of this algorithm have been clearly observed at early stages in the mining process.