scispace - formally typeset
Search or ask a question
Book ChapterDOI

Linear Discriminant Analysis

31 Jul 2003-pp 123-168
About: The article was published on 2003-07-31. It has received 12 citations till now. The article focuses on the topics: Kernel Fisher discriminant analysis & Optimal discriminant analysis.
Citations
More filters
Journal ArticleDOI
TL;DR: The value of mRNA for blood deposition timing is demonstrated and a statistical model for estimating day/night time categories based on molecular biomarkers is introduced, which provides new leads for molecular approaches on time of death estimation using the significantly rhythmic mRNA markers established here.
Abstract: Determining the time a biological trace was left at a scene of crime reflects a crucial aspect of forensic investigations as - if possible - it would permit testing the sample donor's alibi directly from the trace evidence, helping to link (or not) the DNA-identified sample donor with the crime event. However, reliable and robust methodology is lacking thus far. In this study, we assessed the suitability of mRNA for the purpose of estimating blood deposition time, and its added value relative to melatonin and cortisol, two circadian hormones we previously introduced for this purpose. By analysing 21 candidate mRNA markers in blood samples from 12 individuals collected around the clock at 2h intervals for 36h under real-life, controlled conditions, we identified 11 mRNAs with statistically significant expression rhythms. We then used these 11 significantly rhythmic mRNA markers, with and without melatonin and cortisol also analysed in these samples, to establish statistical models for predicting day/night time categories. We found that although in general mRNA-based estimation of time categories was less accurate than hormone-based estimation, the use of three mRNA markers HSPA1B, MKNK2 and PER3 together with melatonin and cortisol generally enhanced the time prediction accuracy relative to the use of the two hormones alone. Our data best support a model that by using these five molecular biomarkers estimates three time categories, i.e. night/early morning, morning/noon, and afternoon/evening with prediction accuracies expressed as AUC values of 0.88, 0.88, and 0.95, respectively. For the first time, we demonstrate the value of mRNA for blood deposition timing and introduce a statistical model for estimating day/night time categories based on molecular biomarkers, which shall be further validated with additional samples in the future. Moreover, our work provides new leads for molecular approaches on time of death estimation using the significantly rhythmic mRNA markers established here.

40 citations


Cites methods from "Linear Discriminant Analysis"

  • ...Besides logistic regression, there is an array of well-established statistics or machine-learning techniques for prediction modelling, such as linear discriminant analysis [56] and support vectormachines [57]....

    [...]

Journal ArticleDOI
TL;DR: The results show that the proposed AE-based model can not only improve the classification accuracy but also be beneficial to solve the problem of False Positive Rate.

39 citations

Journal ArticleDOI
11 Jan 2021
TL;DR: In this paper, the authors proposed an array-based sensing approach for chemical and biological systems, which can provide sensitive and rapid detection of a variety of substrates, including biological and chemical molecules.
Abstract: Chemical sensors play an important role in our understanding of chemical and biological systems, providing sensitive and rapid detection of a variety of substrates. Array-based sensing approaches a...

30 citations

Journal ArticleDOI
TL;DR: In this paper, a data-driven approach for the reliable identification of clay minerals based on spectroscopy and multivariate analyses is presented, which uses Raman and laser-induced breakdown spectroscopies (LIBS) to discriminate geological specimens based on their dominant clay mineralogy.

26 citations

Journal ArticleDOI
Lei Xu1, Jeffrey A. Cruz1, Linda J. Savage1, David Kramer1, Jin Chen1 
TL;DR: A coarse-to-refined model called dynamic filter is developed to identify abnormalities in plant photosynthesis phenotype data by comparing light responses of photosynthesis using a simplified kinetic model of Photosynthesis.
Abstract: Motivation: Plant phenomics, the collection of large-scale plant phenotype data is growing exponentially. The resources have become essential component of modern plant science. Such complex data sets are critical for understanding the mechanisms governing energy intake and storage in plants, and this is essential for improving crop productivity. However, a major issue facing these efforts is the determination of the quality of phenotypic data. Automated methods are needed to identify and characterize alteractions caused by system errors, all of which are difficult to remove in the data collection step, and distinguish them from more interesting cases of altered biological responses. Results: As a step towards solving this problem, we have developed a coarse-to-refined model called Dynamic Filter to identify abnormalities in plant photosynthesis phenotype data by comparing light responses of photosynthesis using a simplified kinetic model of photosynthesis. Dynamic Filter employs an Expectation-Maximization process to adjust the kinetic model in coarse and refined regions to identify both abnormalities and biological outliers. The experimental results show that our algorithm can effectively identify most of the abnormalities in both the real and synthetic datasets.

20 citations