Bienvenue sur le profil Malt de Siying !
Localisation et déplacement
- Localisation
- Paris, France
- Peut travailler dans vos locaux à
-
- Paris et 50km autour
- Antony et 10km autour
Préférences
- Durée de mission
-
- entre 1 et 3 mois
- entre 3 et 6 mois
- ≥ 6 mois
- Secteur d'activité
-
Préfèrerait :
- Biotechnologies
- Centres de recherche
- Education & e-learning
- Edition de logiciels
- Energie
+10 autresPréfèrerait éviter:Automobile
- Taille d'entreprise
-
Préfèrerait :
- 11 - 49 personnes
- 50 - 249 personnes
- 250 - 999 personnes
- 1000 - 4999 personnes
Préfèrerait éviter:- 2 - 10 personnes
- ≥ 5000 personnes
Vérifications
Charte du freelance Malt signée
Consulter la charte
Langues
Catégories
Compétences (20)
- BigData
-
- Data Science
-
1
-
-
-
-
-
- Tous
-
Siying en quelques mots
Au cours de la dernière décennie, j'ai travaillé sur une variété de données du monde réel dans des contextes scientifiques et industriels. Ma force repose sur mes compétences analytiques, qui incluent non seulement la modélisation statistique ou machine learning mais surtout la méthodologie de conception de projet pour obtenir des données utiles.
Spécialités: modélisation statistique de séries temporelles, machine learning, programmation, nettoyage de données, analyse de données, plan d'étude (études cliniques, études d'observation), communication et rapport scientifique.
---
Le tarif affiché sur mon profil est indicatif et est négociable
Portfolio
Expériences
DataVac
Secteur médical
Data scientist - En tant que freelance
- Applied the optical mark recognition (OMR) algorithm for reading medical charts.
- Applied the optical character recognition (OCR) algorithms for numeric and alphabetic input recognition.
Veolia - Veolia
Energie
Data scientist
The objective of the heat load forecast project is to create a decision tool to optimize the operations of heat productions in the power plants. This tool will be able to forecast the heat consumption demand with good precision and guide the decision for boiler operations in the local plants. In this project, I
- led the data analysis and applied a machine-learning algorithm to detect sensor failure periods and make predictions for heat load needs.
- auto-mated the pipeline of data ingestion - data preprocessing - model training - prediction - post-processing - dashboarding in the proof of concept (POC) prototype using Google Cloud Platform.
- collaborated with the developer team on bringing the POC to industrialization.
This product is now used in a city in Europe. It solved the problem that the traditional method cannot obtain heat consumption information in real-time and prevented over and under heat generation.
Veolia - Veolia
Energie
Data scientist
This project aims to provide a user-friendly web interface to evaluate the energy savings provided by the client's service. It had preset several types of statistical and machine learning algorithms and provided interactive templates for user's computational needs. In this project, I
- maintained the R code.
- led the interface layout code refactoring for the second version.
- coordinate with data scientist/developer on Git organizations
- provided expert opinion on statistical concepts for users to understand.
Veolia - Veolia
Environnement
Data scientist
The objective of this project is to monitor annual transaction activities and detect abnormal transaction activities within the client's enterprise. In this project, I
- led the queries from the historical transactions database.
- improved query algorithm for suspicious activity detection
- created dashboard visualization to efficiently capture main points from the results.
INSERM - INSERM
Secteur médical
Statistician/ postdoctoral fellow
The objective of this project is to find the genetic loci associated with FMD, a cardiovascular disease commonly seen in women. In this project, I
- Led the pipeline for genome-wide association analysis which included the quality control genotyped data, genome-wide genotype imputation for untyped regions in case and control samples, quality control for imputed data, association analysis.
- Performed diagnostic analysis on imputed genotype data and bias analysis on association estimations using imputed genotype data.
- Wrote lab guide book for genome-wide analysis procedures
- Presented preliminary results in the international conferences
University of Toronto School of Public Health
Santé & bien-être
Data analyst / postdoctoral fellow
- Worked on a longitudinal environmental health study and a cross-sectional genetic study. The work mainly included:
- Led association analyses on environmental exposures and their influences on neurobehavioral outcomes in Mexican children.
- Led the analysis of the genetic modifying effect of health outcomes in children exposed to environmental lead pollution.
- Led the study on environmental fluoride exposure and oral health outcome in Thai children.
- Corrected massive sample mislabeling using genetic statistical algorithms.
- Works were either published in the international journal or presented in international conferences
Genetic and epigenetic study
- Conducted an association analysis on DNA methylation profiling and circulating lipoprotein A levels in French Canadian families.
- Conducted a Monte Carlo simulation for genetic association test with left-censored distribution in the trait.
3 recommandations externes
Consultez les recommandations qu'a reçues Siying