Data Analyst Accounting Manager

I am a driven and passionate Data Analyst with * years’ experience delivering

customer focused products by leveraging the power of data science and business intelligence to achieve customer retention and revenue upliftment.. P R O F E S S I O N A L O B J E C T I V E

7100 Almeda Rd, Apt 624 Houston, Tx 77054.

Phone: 832-***-****

Email: *******@*****.***

https://www.linkedin.com/in/williamchavula/

C O N T A C T D E T A I L S

Oracle SQL

Postgres SQL

Python

Microsoft Excel

Tableau

R E L A T E D S K I L L S

W I L L I AM R . C H A V U L A

Power BI

Project Management

Time Management

Reporting

E D U C A T I O N

University of Malawi, Chancellor College. Zomba, Malawi 2007 - 2010

Bachelor of Science (BSc)

Major: Computer Science and Statistics

Machine Learning

Python for Data Science

Python Programming

Machine learning Model Deployment

Strategic Management

Bioinformatics (Computational Biology)

Open Online courses (MOOCs)

WORK E X P E R I E N C E

W I L L I AM R . C H A V U L A

Employed Customer analytics and developed models that resulted in increased data activity of price elastic customers by over 50%, resulting in a 20% increase in data revenue.

Evaluated revenue, customer acquisition, and pricing models against set business goals and targets, making necessary adjustments in the input variables.

Designed and implemented innovative mobile data packages in collaboration with analytics partners and product management team. Translated high-level product design concepts to practical system implementation ensuring correct functionality and charging with minimal time to market.

Compiled Business Intelligence reports on revenue, minutes of use, and subscriber metrics to create a comprehensive picture of business performance. Managed and implemented various projects within the business, which delivered enhanced customer analytics and intelligent pricing.

·Trained, led, and supervised a team of network Surveillance Engineers. Compiled Root Cause Analysis (RCA) reports which became the standardized procedure for preventing and rectifying critical network failures. Designed and implemented incident escalation matrices and fault dictionaries reducing network downtime by over 40%.

Contributed to the implementation of network surveillance policies, procedures and processes ensuring quick and efficient handling of network incidents.

TELEKOM NETWORKS MALAWI (TNM)

Business Analyst, Blantyre, Malawi (02/2015-09/2019) Fault Management Supervisor, Lilongwe, Malawi (06/ 2011– 01/2015) D A T A A N A L Y S I S P R O J E C T S

W I L L I AM R . C H A V U L A

The time of day is highly correlated to the crimes (both type and amount committed. In other words, the time of day determines both the amount and type of crime committed. We observed that hours from 12 noon to midnight contribute to 65% of crime occurrences.

The location, given by the longitude and latitude, and the location type also determine the type of crime that occur in that particular area. We observed that most crimes occur in residential/ home areas as well as highways, streets, and alleys.

For our particular task, a useful metric to evaluate our model was a high recall.We were interested in a model that is able to minimize false negatives. This would help our efforts to curb crime. It is better to have a false alarm than no alarm which turns out to be catastrophic.

AUSTIN TX PD CRIME CLASSIFICATION, November 2019

DESCRIPTION:

Explored crime data from Austin Police department with the aim of understanding the relationships between features. The analysis revealed: Afterwards,I built a model using the GradientBoostingClassifier from scikit-learn. I evaluated the model based on how well it was able to reduce the number of false negatives as our model should be able to classify crime to a high degree without missing or misclassifying any.

D A T A A N A L Y S I S P R O J E C T S

W I L L I AM R . C H A V U L A

First round of exploratory data analysis didn’t reveal any interesting insights regarding key features that influence accidents.

I discovered insights such as, most accidents happen during the day, in the morning. Severity 2 (in a range of 0 to 4) is the most common accident severity. Features such as wind speed, Temperature, Humidity do not play any significant role with respect to severity.

Following these findings my next step is to explore correlations between the severity feature and the other features. The problem I would like to solve is finding the most efficient and accurate way of calculating correlation between categorical variables and continuous variables, and categorical variables and categorical variables. I believe these will reveal intricate relationships that the traditional EDA didn’t reveal and will provide insights for building the model. US ACCIDENTS - A COUNTRYWIDE TRAFFIC ACCIDENT DATASET. 01/2020 – current

DESCRIPTION

Exploring factors that are influential in causing accidents. Dataset contains features such as: Temperature, Wind Speed, Wind Chill, Humidity, Accident Description, Time of the accident and Accident Severity, etc. The goal is to develop a model that will predict accident severity using the features given. A T T R I B U T E S A N D H O B B I E S

Ardent learner

Critical Thinker

Goal Oriented

Planner and Organized

Team Player

Self-Starter

Hiking

Reading

Travel

Art

R E F E R E E S

W I L L I AM R . C H A V U L A

Mrs. Chikondi Mkoko,

Income Accounting Manager,

Telekom Networks Malawi (TNM),

P.O. Box 3039,

Livingstone Towers,

Glyn Jones Road,

Blantyre, Malawi.

Email: ********.*****@***.**.**

Tel : +265*********

Mr Gift Medi,

Head of ICT (CIO),

Medical Aid Society of Malawi (MASM),

P.O. Box 1254,

MASM House,

Lower Scalter Road,

Blantyre, Malawi.

Email: *****@****.**

Tel: +265*********

Contact this candidate