Hala Helmi

Resume posted by HalaHelmi in IT.
Desired salary: $30,000.00
Desired position type: Full-Time
Location: New Valley Governorate Egypt

Contact HalaHelmi


Data Analyst with 6 years of experience in several sectors. Experienced in preparing detailed documents and reports while managing complex internal and external data analysis responsibilities. Proactive, results-driven professional with a solid track record in data collection, cleansing and visualization that optimize data management, capturing, delivery and quality.




University of Nottingham


Doctor of Philosophy (PhD), Computer Science, 2010 – 2014

Thesis title is Developing Methods for Machine Learning Algorithms Using Automated Feature and Sample Selection.


University of Nottingham


Master of Science (MSc), Management of Information Technology,

Computer Science School, 2008 – 2009 Grade: Merit


Dissertation title is “Studying the Performance of Monitoring the Routing Using SNMP on OSPF Protocol with Down State Neighbour: Using OPNET”



King Abd Al Aziz University


Bachelor of Science (BS), Management Information Systems, 2003 – 2007

Grade: 4.60/5


Dissertation title is “Examine the effectiveness and efficiency of information system used in banks to increase the performance of staff in banks”.


  • 3rd Year Project: Establishment a database for the Ministry education, Branch of Islamic study in Makkah, Saudi Arabia.
  • 4th year Project: Analysis and redesign of an information system of SDAFCO company SDAFCO, Jeddah, Saudi Arabia.




Data Scientist at TA Telecom

January 2016 – Present

Working as Data Scientist for MegaKheir which is a charity arm of TA Telecom. It is a mobile donation application whereby users send text messages via their mobile phones to donate.

As Customer analytics my role including:

  • Customer Segmentation and Profiling (RFM, Bayesian modelling),
  • Customer Lifetime Value (segmentation, Pareto/NBD),
  • Churn detection (survival analysis, logistic regression),
  • Product Recommendation (association rules mining), designing a Single Customer View.
  • Developed a number of tests to increase the ROI of the NGOs campaign using propensity model (logistic regression) which has led to immediate cost savings and is expected to increase the lifetime value of supporters considerably.


Credit Risk analyst at Score

January 2015 – September 2015

As a freelance member in the Credit Risk analysis team I was responsible to identifying, assessing, measuring, and monitoring credit risks and ensuring appropriate risk controls and responses with high accuracy were applied. In addition to that, provide predictive modeling support analyzing customer data, develop/optimize strategies for loan pricing.



Data Analyst at  Bey2ollak.com

July 2014  – January 2015 (7 months)

Bey2ollak is based on a very simple idea; a community based traffic information service aiming to keep track of the status roads users give live feedback on how the traffic is.


  • Find meaningful traffic patterns from historical dataset in order to predict traffic flow volumes that can be expected on the road during specific periods by clustering traffic pattern, cognizance should be taken of the fact that traffic volumes change considerably at each point in time. As well as, calculation of Traffic Growth Rates of traffic volumes.


  • There are three cyclical variations that are of particular interest: Hourly pattern: The way traffic flow characteristics varies throughout the day and night; Daily Pattern: The day-to-day variation throughout the week; and Monthly and yearly pattern.


PhD Researcher at University of Nottingham


May 2010 – June 2014 (4 years 2 months)


  • My research mainly focused in developing data mining techniques by investigating the use of semi-supervised methods to identify new classes for breast cancer patients for clinical practice meaningful classification using Nottingham Tenovus Breast Cancer dataset.
  • Developing Machine Learning Algorithms to Automat Feature and Sample Selection. The need of this raised up after coordinated with doctors for data cleaning and variables selection to ensure data clinical meaningful and analyzable.
  • Conducted data analysis using logistical model, LDA, SVM, KNN, tree classification and random clustering method to identify high breast cancer risk patient and improved the accuracy (C-scores) by 28%.
  • Four papers related to this project submitted and published.


  • MS Office Suite Experienced
  • SPSS
  • SAS Intermediate
  • Intermediate
  • Tableau Intermediate
  • Programming Languages:
  • R Experienced
  • Python Experienced
  • SQL Experienced
  • Java Intermediate

Spoken Languages

    Arabic, English