7 research outputs found
To Tune or not to Tune: Hyperparameter Influence on the Learning Curve
No full textA learning curve displays the measure of accuracy/error on test data of a machine learning algorithm trained on different amounts of training data. They can be modeled by parametric curve models that help predict accuracy improvement through curve extrapolation methods. However, these learning curves have only been mainly generated from default learning algorithms. Research into tuning the machine learning algorithm and its effect on the learning curve has not been adequately researched. This research aims to look at the influence of hyperparameter tuning on the learning curve. This regards not only how the learning curve shape changes in general but also how different parametric models are affected when a learner undergoes tuning. We experiment with the decision tree and KNeighbors classifier which undergo significant hyperparameter tuning. We find that the tuned learner performs marginally better than the default learner for anchors past 25\% of the data for the majority of the tested datasets. We also observe that the tuned learner displays a smoothing behaviour that makes ill-behaved curves more well-behaved. In terms of the curve fitting, the tuned learners do not uncover any curve models nor does it show any statistical significance, and instead performs very similarly to the default learners. CSE3000 Research ProjectComputer Science and Engineerin
Recommended from our members
To Impute or Not To Impute in Untargeted Metabolomics─That is the Compositional Question.
No full textUntargeted metabolomics often produce large datasets with missing values. These missing values are derived from biological or technical factors and can undermine statistical analyses and lead to biased biological interpretations. Imputation methods, such as k-Nearest Neighbors (kNN) and Random Forest (RF) regression, are commonly used, but their effects vary depending on the type of missing data, e.g., Missing Completely At Random (MCAR) and Missing Not At Random (MNAR). Here, we determined the impacts of degree and type of missing data on the accuracy of kNN and RF imputation using two datasets: a targeted metabolomic dataset with spiked-in standards and an untargeted metabolomic dataset. We also assessed the effect of compositional data approaches (CoDA), such as the centered log-ratio (CLR) transform, on data interpretation since these methods are increasingly being used in metabolomics. Overall, we found that kNN and RF performed more accurately when the proportion of missing data across samples for a metabolic feature was low. However, these imputations could not handle MNAR data and generated wildly inflated or imputed values where none should exist. Furthermore, we show that the proportion of missing values had a strong impact on the accuracy of imputation, which affected the interpretation of the results. Our results suggest imputation should be used with extreme caution even with modest levels of missing data and especially when the type of missingness is unknown
Mobipedia: Mobile Applications Linked Data
No full text14th International Semantic Web Conference (ISWC 2015), October 2015We present Mobipedia, an integrated knowledge base with
information about 1 million mobile applications (apps) such as their category, meta-data (author, reviews, rating, release date), permissions and
libraries used, and similar apps. The goal of Mobipedia is to integrate
unstructured and semi-structured data about mobile apps from publicly
available data sources and publish it as Linked Data using RDF. We describe the extraction process for facts, access mechanisms to the knowledge base, and an overview of applications facilitated by Mobipedia.This research work has been supported by RADICLE project
CNS-1059436, CNS-1212943, CNS-1118127 and CNS-1450768, CICYT project TIN2013-
46238-C4-4-R and DGA FSE, U.S. National Science Foundation awards 0910838 and
1228198.https://robertoyus.com/publication/iswc2-2015
Fintech founders: inspiring tales from the entrepreneurs that are changing finance/ Agustín Rubini.
No full textIncludes index.In English."Over 70 in-depth interviews of Fintech Founders provide lessons from some of the most successful fintech entrepreneurs that will help you understand the challenges and opportunities of applying technology and collaboration to solve some key problems of the financial services industry. This book is for entrepreneurs, for people working inside of large organizations and everyone in between who is interested to learn the secrets of successful entrepreneurs. In this advice-filled resource, Rubini gathers advice that comes from a diverse range of financial services niches including financing, banking, payments, wealth management, insurance, and cryptocurrencies, to help you harness the insights of thought leaders. Those working inside the financial services industry and those interested in working in or starting up businesses in financial services will learn valuable lessons on how to take an idea forward, how to find the right business founders, how to seek funding, how to learn from initial mistakes, and how to define and reposition your business model. Rubini also inquires into the future of fintech and uncovers provoking and insightful predictions."--Frontmatter -- About the Author -- Foreword -- Contents -- Preface -- Part 1: Financing Fintechs -- Introduction -- Chapter 1. Henrique Dubugras -- Chapter 2. Renaud Laplanche -- Chapter 3. Levi King -- Chapter 4. Sam Graziano -- Chapter 5. Michael Garrity, Paul Sehr, Casper Wong -- Chapter 6. Sergio Furio -- Chapter 7. Alejandro Cosentino -- Chapter 8. Christoph Rieche -- Chapter 9. Conrad Ford -- Chapter 10. Gamal Moukabary -- Chapter 11. Geetansh Bamania -- Chapter 12. Kelvin Teo -- Chapter 13. Harshvardhan Lunia -- Chapter 14. Simon Loong -- Part 2: Banking and Savings Fintechs -- Introduction -- Chapter 15. Anthony Thomson -- Chapter 16. Nick Ogden -- Chapter 17. Norris Koppel -- Chapter 18. Ricky Knox -- Chapter 19. Mutaz Qubbaj -- Chapter 20. Matthias Kröner -- Chapter 21. Tamaz Georgadze -- Chapter 22. Dr. Yassin Hankir -- Chapter 23. Brett King -- Chapter 24. Pierpaolo Barbieri -- Part 3: Payments Fintechs -- Introduction -- Chapter 25. Mike Massaro -- Chapter 26. Patrick Postrehovsky -- Chapter 27. Sami Louali -- Chapter 28. Elizabeth Rossiello -- Chapter 29. Brett Meyers -- Chapter 30. Christo Georgiev -- Chapter 31. Jacob de Geer -- Chapter 32. Arpit Gupta -- Chapter 33. Wong Joo Seng -- Chapter 34. Prajit Nanu -- Part 4: SME-Specific Fintechs -- Introduction -- Chapter 35. Gert Sylvest -- Chapter 36. Gordon Trouncer Downes -- Chapter 37. Sebastián Cadenas -- Chapter 38. Joel Perlman -- Chapter 39. Tim Fouracre -- Chapter 40. Nicolas Reboud, Raphaël Simon -- Chapter 41. Johan Lorenzen -- Chapter 42. Sean Yu -- Part 5: Investment Fintechs -- Introduction -- Chapter 43. Aaron Klein -- Chapter 44. Mazy Dar -- Chapter 45. John Fawcett -- Chapter 46. Facundo Garreton -- Chapter 47. Gonçalo de Vasconcelos -- Chapter 48. Yoni Assia -- Chapter 49. Adam Leonard -- Chapter 50. Barry Freeman -- Chapter 51. Mike Kayamori -- Part 6: Insurance Fintechs -- Introduction -- Chapter 52. Karn Saroya -- Chapter 53. Tim Attia -- Chapter 54. Michael Serbinis -- Chapter 55. Barry McCarthy -- Chapter 56. Dr. Christopher Oster -- Chapter 57. Talal Bayaa -- Chapter 58. Gustaf Agartson -- Part 7: Data and Analytics Fintechs -- Introduction -- Chapter 59. Stephane Dubois -- Chapter 60. Zor Gorelov -- Chapter 61. Gunnar Carlsson and Gurjeet Singh -- Chapter 62. Walter Alini, Daniel Moisset, Javier Mansilla, Juan Chacon -- Chapter 63. Steve Kirsch and Marten Nelson -- Chapter 64 .Matthew Hodgson -- Chapter 65. Jonathan Epstein -- Part 8: Support Fintechs -- Introduction -- Chapter 66. Steve Polsky -- Chapter 67. Stephen Ufford -- Chapter 68. Sebastian Stranieri -- Chapter 69. Niall Twomey -- Chapter 70. Owen Hall and Vikas Tripathi -- Chapter 71. Bill Safran -- Chapter 72. Raz Abramov -- Chapter 73. William Wei -- Join our newsletter -- Index1 online resource (XVIII, 579 pages)
A resource to empirically establish drug exposure records directly from untargeted metabolomics data
No full textDespite extensive efforts, extracting medication exposure information from clinical records remains challenging. To complement this approach, here we show the Global Natural Product Social Molecular Networking (GNPS) Drug Library, a tandem mass spectrometry (MS/MS) based resource designed for drug screening with untargeted metabolomics. This resource integrates MS/MS references of drugs and their metabolites/analogs with standardized vocabularies on their exposure sources, pharmacologic classes, therapeutic indications, and mechanisms of action. It enables direct analysis of drug exposure and metabolism from untargeted metabolomics data, supporting flexible summarization at multiple ontology levels to align with different research goals. We demonstrate its application by stratifying participants in a human immunodeficiency virus (HIV) cohort based on detected drug exposures. We uncover drug-associated alterations in microbiota-derived N-acyl lipids that are not captured when stratifying by self-reported medication use. Overall, GNPS Drug Library provides a scalable resource for empirical drug screening in clinical, nutritional, environmental, and other research disciplines, facilitating insights into the ecological and health consequences of drug exposures. While not intended for immediate clinical decision-making, it supports data-driven exploration of drug exposures where traditional records are limited or unreliable.This project was enabled in part by the Alzheimer’s Gut Microbiome Project (AGMP) and the Data Infrastructure and Molecular Atlas for AD: Connection Exposome, Gut Microbiome, and Metabolome supplement funded wholly or in part by the following grants thereto: 1U19AG063744 and 3U19AG063744-04S1 and awarded to Dr. Kaddurah-Daouk at Duke University in partnership with multiple academic institutions. As such, the investigators within the AGMP and the Exposome Supplement, not listed specifically in this publication’s author list, provided data along with their pre-processing and prepared it for analysis, but did not participate in analysis or writing of this manuscript. A listing of AGMP Investigators can be found at https://alzheimergut.org/meet-the-team/. A complete listing of ADMC investigators can be found at: https://sites.duke.edu/adnimetab/team/. We also thank the support by NIH for the Maternal and Pediatric Precision in Therapeutics project P50HD106463, the development of tools for structure elucidation R01DK136117, and the Collaborative Microbial Metabolite Center U24DK133658. The HIV Neurobehavioral Research Center (HNRC) is supported by Center award P30MH062512 from NIMH. This research was supported in part by the Intramural Research Program of the NIH, National Institute of Environmental Health Sciences (ZIC ES103363). H.N.Z. was supported by the National Institute of Environmental Health Sciences of the National Institutes of Health under Award Number K99ES037746. C.B. was supported by the Czech Academy of Sciences PPLZ fellowship number L200552251. V.C.L. is supported by Fonds de recherche du Québec - Santé (FRQS) Postdoctoral fellowship (335368). N.E.A was supported in part by the National Center for Complementary and Integrative Health of the NIH under award number F32AT011475. A.M.C.-R. and P.C.D. were supported by the Gordon and Betty Moore Foundation grant GBMF12120. M.R. was supported by the NIH grant R37 AI126277. T.P. was supported by the Czech Science Foundation (GA CR) grant 21-11563 M and by the European Union’s Horizon 2020 research and innovation programme under Marie Skłodowska-Curie grant agreement No. 891397. L.C., R.G.-S., and P.G.-F. were supported by the Spanish Ministry of Science (PID2022-139446OB-C21 and PID2022-139446OB-C22). L.C. acknowledges the support from the Economy and Knowledge Department of the Catalan Government through Consolidated Research Group (ICRA-TECH 2021 SGR 01283), as well as from the CERCA programme. W.B. acknowledges support by the Research Foundation–Flanders (FWO G0AHY25N).Peer reviewe
The microbiome diversifies N-acyl lipid pools - including short-chain fatty acid-derived compounds.
No full text N-acyl lipids are important mediators of several biological processes including immune function and stress response. To enhance the detection of N-acyl lipids with untargeted mass spectrometry-based metabolomics, we created a reference spectral library retrieving N-acyl lipid patterns from 2,700 public datasets, identifying 851 N-acyl lipids that were detected 356,542 times. 777 are not documented in lipid structural databases, with 18% of these derived from short-chain fatty acids and found in the digestive tract and other organs. Their levels varied with diet, microbial colonization, and in people living with diabetes. We used the library to link microbial N-acyl lipids, including histamine and polyamine conjugates, to HIV status and cognitive impairment. This resource will enhance the annotation of these compounds in future studies to further the understanding of their roles in health and disease and highlight the value of large-scale untargeted metabolomics data for metabolite discovery. </p
The microbiome diversifies long- to short-chain fatty acid-derived N-acyl lipids
No full textN-Acyl lipids are important mediators of several biological processes including immune function and stress response. To enhance the detection of N-acyl lipids with untargeted mass spectrometry-based metabolomics, we created a reference spectral library retrieving N-acyl lipid patterns from 2,700 public datasets, identifying 851 N-acyl lipids that were detected 356,542 times. 777 are not documented in lipid structural databases, with 18% of these derived from short-chain fatty acids and found in the digestive tract and other organs. Their levels varied with diet and microbial colonization and in people living with diabetes. We used the library to link microbial N-acyl lipids, including histamine and polyamine conjugates, to HIV status and cognitive impairment. This resource will enhance the annotation of these compounds in future studies to further the understanding of their roles in health and disease and to highlight the value of large-scale untargeted metabolomics data for metabolite discovery.</p
