Wenjun Guo’s research while affiliated with Chinese University of Hong Kong and other places

What is this page?


This page lists works of an author who doesn't have a ResearchGate profile or hasn't added the works to their profile yet. It is automatically generated from public (personal) data to further our legitimate goal of comprehensive and accurate scientific recordkeeping. If you are this author and want this page removed, please let us know.

Publications (4)


A Survey on Dimension Reduction Algorithms in Big Data Visualization
  • Chapter

May 2020

·

126 Reads

·

9 Citations

Lecture Notes of the Institute for Computer Sciences

Zheng Sun

·

Weiqing Xing

·

Wenjun Guo

·

[...]

·

Shenghui Cheng

In practical applications, the data set we deal with is typically high dimensional, which not only affects training speed but also makes it difficult for people to analyze and understand. It is known as “the curse of dimensionality”. Therefore, dimensionality reduction plays a key role in the multidimensional data analysis. It can improve the performance of the model and assist people in understanding the structure of data. These methods are widely used in financial field, medical field e.g. adverse drug reactions and so on. In this paper, we present a number of dimension reduction algorithms and compare their strengths and shortcomings. For more details about these algorithms, please visit our Dagoo platform via www.dagoovis.com.


Figure 2: Data processing structure. Framework for the information acquisition from temporal event data.
Figure 3: The visual expression shows both multivariate data features and temporal changes in one view.
BubbleUp: Toward Better Analysis for Temporal Event Data
  • Conference Paper
  • Full-text available

October 2018

·

204 Reads

Temporal event data such as diagnosis records of multiple demen- tia patients include both multivariate data features and temporal changes. Analyzing temporal event data can reveal changing pat- terns by time or by correlations between target data and others. It is challenging but to explore visually both of multivariate data features and temporal changes in one view. In this study, we present a novel visualization system named BubbleUp which can explore temporal event databased on machine learning methods to con rm the correlations among the data and visualizes the changing patterns by time. The usage of BubbleUp visualization system can be divided into four steps; Overall distribution, detail view, correlation base on similarity, and prediction. We evaluate the usage and effectiveness of BubbleUp visualization system through the usage scenarios.

Download

Data-driven dementia diagnosis record visualization system

August 2017

·

11 Reads

In this study, we propose 'Dementia Tracker' which is a visualization system with 21,094 dementia records for 8 years. This system makes it easy to understand complex dementia record data. In addition, the patient's own record is not only well-read, but also can be compared with people who have similar degree of dementia through the group filter function. Therefore, the current dementia situation can be understood more easily, and the future dementia can be predicted and prevented in advance.


Garden Agua: Three-Dimensional tangible display enabled by arranged water jet

November 2013

·

18 Reads

·

4 Citations

Levitation refers to the process of floating in the air in a stable position without physical contact points against gravity. This was a mysterious phenomenon that many have wanted to experience for many years. Such wishes are expressed in the Bible, for instance when Jesus crossed the Red Sea, and when Peter wanted to walk on water with Jesus. Such phenomena, floating in the air and walking on water, were nothing short of miracles to people. Garden Agua was devised to realize these human aspirations.

Citations (2)


... Algorithms like principal component analysis (PCA) help to achieve dimensionality reduction by finding the prominent features of the data. (30,31,32) In adding, CT scan-derived brain image features have a lot of dimensions which can make it hard to train a model without falling into overfitting issues. ...

Reference:

Detecting hemorrhagic stroke from computed tomographic scans using machine learning models comparison
A Survey on Dimension Reduction Algorithms in Big Data Visualization
  • Citing Chapter
  • May 2020

Lecture Notes of the Institute for Computer Sciences

... The user in this case is passive, and unable to dictate or influence the output. Another type of shape-changing prototype that is excluded is Guo et al's Garden Agua [27] despite being described as shape-changing display in the literature-as it deals only with moveable solid objects and not surface deformation. The same premise also applies to Ariel Tunes [3] due to the modular and limited nature of its current form-based output. ...

Garden Agua: Three-Dimensional tangible display enabled by arranged water jet
  • Citing Conference Paper
  • November 2013