$$$$
{{ $t($store.state.user.experience_value_in_dollars) }}
Expert
{{ $t($store.state.user.experience_search_name) }}
0
jobs
Alex Dolia
,
, United Kingdom
Experience
Other titles
Skills
I'm offering
Markets
United Kingdom
Language
English
Fluently
My experience
2017 - ?
job
Lead Data Scientist
Deep Intellect.
2016 - 2016
job
Senior Data Scientist
Capita Employee Benefits.
• Developed several models for a number of clients, including corporate travel agencies, insurance companies, corporate communication organisations, employee benefits and HR analytics.
• Developed several models for a corporate travel agency based on survival analysis, logistic regression and a generalised linear model with a Poisson distribution. The outcome shows correlations exist between travel behaviour and several work performance indicators.
• In charge of designing mathematical models and computational solutions such as machine learning algorithms.
• Managed the implementation of a probabilistic model for identifying unallocated assets for clients in the insurance industry
• Organized a strategic governance workshop for a telecommunication company which included commercial insights from demographics cuts to benefits strategies implementation.
• Develop R&D code for Capita Group using Python, R, Tabular and SQL.
• Coached and supervised junior members of staff on technical issues. These members had a dotted line to me.
• Developed several models for a corporate travel agency based on survival analysis, logistic regression and a generalised linear model with a Poisson distribution. The outcome shows correlations exist between travel behaviour and several work performance indicators.
• In charge of designing mathematical models and computational solutions such as machine learning algorithms.
• Managed the implementation of a probabilistic model for identifying unallocated assets for clients in the insurance industry
• Organized a strategic governance workshop for a telecommunication company which included commercial insights from demographics cuts to benefits strategies implementation.
• Develop R&D code for Capita Group using Python, R, Tabular and SQL.
• Coached and supervised junior members of staff on technical issues. These members had a dotted line to me.
Sql, Python, Machine learning, R, Analytics, Corporate communication, Algorithms, Implementation, Insurance, Workshop, ME
2014 - 2016
job
Data Scientist
Big Data Partnership.
• Researched and developed a unique way of recommending brands to consumers based on predicting gender, ethnicity and location. Demographics of users were identified from twitter data. Solution included support vector machines, logistic regression and random forest to create the models using feature extraction techniques such as bag of words and word2vec. Identified demographics of the users applying Natural Language Processing to twitter data such as flexible regular expressions.
• Developed streaming software to collect reliable training dataset from twitter using twitter API and Natural Language Processing.
• Developed a solution to model NHS data using Survival analysis (Cox proportional hazard model and Kaplan-Meier estimator) and Bayesian Network modelling to model the probability of developing stroke or heart attack based on demographics, diet, drinking habits and level of physical exercises.
• Developed convolution neural network for human/non human classification of twitter profile images and car maker detection using tensor flow
• Developed code using Python, R, Scala and Spark
• Developed streaming software to collect reliable training dataset from twitter using twitter API and Natural Language Processing.
• Developed a solution to model NHS data using Survival analysis (Cox proportional hazard model and Kaplan-Meier estimator) and Bayesian Network modelling to model the probability of developing stroke or heart attack based on demographics, diet, drinking habits and level of physical exercises.
• Developed convolution neural network for human/non human classification of twitter profile images and car maker detection using tensor flow
• Developed code using Python, R, Scala and Spark
Python, API, Training, R, Scala, Spark, Network, Natural, Twitter, Support, Software, Streaming, Feature, Processing, Twitter api
2014 - 2014
job
Senior Statistician
Imperial College NHS Trust.
• Utilized a number of statistical modelling techniques such as trend detection using generalized linear models and distribution change detection. These were applied to incidents that occurred within the NHS. Applied several data manipulation and cleaning methods to create a good dataset.
• Produced several reports and presented the results to a non-specialist audience.
• SAS code development was used in my solution.
• Produced several reports and presented the results to a non-specialist audience.
• SAS code development was used in my solution.
SAS, Development
2013 - 2013
job
Decision Science Analyst
Wonga.
• Developed scorecards for probability of default estimation and data extraction methods for Wonga Little Loan product using R, SQL and such techniques as logistic regression, decision trees and pmml files.
• Developed methods for fraud detections using unsupervised learning, customer data and credit bureau information.
• Identified behavioural patterns and classified according to risk scenarios.
• Produced evaluations of the quality of detection algorithms.
• Performed data mining using SQL and R code for customer segmentation analysis.
• Developed SQL code for monitoring of decision workflow.
• Experience with DecisionLogic, Callcredit, Experian data and report writing.
• Developed methods for fraud detections using unsupervised learning, customer data and credit bureau information.
• Identified behavioural patterns and classified according to risk scenarios.
• Produced evaluations of the quality of detection algorithms.
• Performed data mining using SQL and R code for customer segmentation analysis.
• Developed SQL code for monitoring of decision workflow.
• Experience with DecisionLogic, Callcredit, Experian data and report writing.
Sql, Writing, Data mining, R, Algorithms, Workflow, Analyst, Monitoring, Science, Patterns
2012 - 2013
job
Research Fellow - Centre for Risk Research
University of Southampton.
• Applied a two-stage Kalman filter, developed Particle Filter and multinomial/conditional logit model for Hong Kong horse racing data. While a forecast target of the finishing order of race participants is appropriate in that it is highly explanatory of past race results, found that it is both too noisy and too dependent on external factors for it to be an effective target. Instead, proposed to forecast future values of attributes of competitors that have a high predictive power of race results. The project funded by the horse betting company.
• Developed survival analysis to identify trading discipline characteristics of traders of a UK-based spread trading company (e.g., gender, holding periods, inclination towards the disposition effect etc.) Associated with greater survivorship during the dramatic fall in prices in Sept 2008 (i.e. associated with their tendency to continue to trade after the crisis.
• Developed survival analysis to identify trading discipline characteristics of traders of a UK-based spread trading company (e.g., gender, holding periods, inclination towards the disposition effect etc.) Associated with greater survivorship during the dramatic fall in prices in Sept 2008 (i.e. associated with their tendency to continue to trade after the crisis.
Research, It, Power, Forecast
2010 - 2012
job
Quantitative Analyst
unknown.
• Developed tools for algorithm trading using exponential smoothing, a relative strength index, logistic regression and support vector machines.
• Developed a robust linear regression model and tested its viability as an equity trading strategy.
• Developed a robust linear regression model and tested its viability as an equity trading strategy.
Support, Analyst
2009 - 2010
job
Senior Statistician
Natural Environment Research Council.
Project title: Case study II - Climate modelling
• Developed a case study using the intermediate complexity climate model GENIE-1 that predicts the global temperature on the Earth by changing 8 parameters.
• Developed several algorithms based on Latin hypercube, Sobol sequences and emulators (Gaussian processes).
• Developed a case study using the intermediate complexity climate model GENIE-1 that predicts the global temperature on the Earth by changing 8 parameters.
• Developed several algorithms based on Latin hypercube, Sobol sequences and emulators (Gaussian processes).
Algorithms, Processes
2007 - 2009
job
Research Fellow
Southampton Statistical Science Research Institute.
Southampton, UK
Project title: Adaptive Design of Experiments for Spatio-temporal Models
• Proposed a Bayesian adaptive design of experiment method for the Gaussian processes and state-based dynamical (linear dynamic) model of ozone pollution.
• Compared Bayesian and Maximum Likelihood Methods for Fitting Space-time Hierarchical Models and discovered that a Bayesian approach based on Gibbs sampling was less sensitive to initial conditions.
Project title: Adaptive Design of Experiments for Spatio-temporal Models
• Proposed a Bayesian adaptive design of experiment method for the Gaussian processes and state-based dynamical (linear dynamic) model of ozone pollution.
• Compared Bayesian and Maximum Likelihood Methods for Fitting Space-time Hierarchical Models and discovered that a Bayesian approach based on Gibbs sampling was less sensitive to initial conditions.
Design, Research, Less, Processes
2002 - 2003
job
Research Fellow - Psychology Department
University of Glasgow.
Project title: New Approach to Human Stereo-Motion Integration
• Developed methodology for analysing motion in depth in visual cortex by mathematical modelling and computer simulations.
• Developed feature extraction and kernel methods (e.g., Parzen Window) to recognise the different styles of human motion
• Developed methodology for analysing motion in depth in visual cortex by mathematical modelling and computer simulations.
• Developed feature extraction and kernel methods (e.g., Parzen Window) to recognise the different styles of human motion
Research, Integration, Feature
1994 - 2002
job
Research Fellow - Radar Remote Sensing and Medical Systems Research Centre
National Aerospace University.
• Developed neural networks for locally adaptive radar image classification and filtering.
• Developed local Radon transforms to extract curves and short line segments.
• Design Fourier-based fast algorithms for radar image and sonar signal processing.
• Implementing these algorithms in Pascal and Matlab.
• Developed local Radon transforms to extract curves and short line segments.
• Design Fourier-based fast algorithms for radar image and sonar signal processing.
• Implementing these algorithms in Pascal and Matlab.
Design, Research, Matlab, Algorithms, Neural networks, Signal processing, Processing
My education
?
-
2009
University of Southampton
MSc, Statistics
MSc, Statistics
?
-
2001
National Aerospace University
Doctorate, Machine Learning and Image/Signal Processing
Doctorate, Machine Learning and Image/Signal Processing
?
-
1994
Kharkov Aviation Institute
MSc, Computer Science
MSc, Computer Science
Alex' reviews
Alex has not received any reviews on Worksome.
Contact Alex Dolia
Worksome removes the expensive intermediaries and gives you direct contact with relevant talent.
Create a login and get the opportunity to write to Alex directly in Worksome.
38100+ qualified freelancers
are ready to help you
Tell us what you need help with
and get specific bids from skilled talent in Denmark