Date

 

 

 

Hiring Agent Name

Title

Company Name

Address

City/State/Zip Code

 

Dear__________________:

 

I am currently seeking a challenging career opportunity in a <insert job title/position> capacity and am submitting my resume for your review.  In advance, thank you for your time and consideration.

 

As indicated in the accompanying resume, I bring to you exceptional academic credentials, proficiency in artificial intelligence, and an exemplary blend of talents in conducting research, performing statistical analysis, medical data mining, and anomaly detection. To complement these qualifications, I offer outstanding documentation, quality assurance, and client relations expertise.

 

As a proactive contributor to your organization, you will find me to be a self-motivated professional committed to supporting you in achieving your objectives through diligence, versatility, and polished troubleshooting skills. I am confident that I could be a valuable asset to your company, and look forward to interviewing with you in the near future.

 

Sincerely,

 

 

 

Yyyyyy x. yyyyyy

 

 

 

Profile

 

Excel in using machine learning and data mining techniques such as Support Vector Machines (SVM), neural networks, decision trees, genetic algorithms and clustering techniques such as self-organizing maps and K-Nearest Neighbor.  Proven skills in medical data mining; able to develop logical data mining models to facilitate diagnosis from patient databases. Significant talents in pattern recognition of chemical sensor signatures with specialization in using machine learning techniques to detect chemical warfare agents and toxic industrial chemicals. Proficiency in statistical analysis and signal processing of chemical sensor signals.

Considerable skills in anomaly detection and activity recognition; demonstrated capabilities in detecting abnormal activities from disparate sensor data streams, defining speech recognition features from acoustic sensors, and interpreting 3D accelerometer readings.

 

Selected Achievements:

 

• Designed and developed features for chemical detection which measurably improved performance of multi-class SVM models.

         Enhanced performance and stability of chemical detection models through feature selection via genetic algorithms and parameter optimization.

         Expedited training time and elevated chemical detection performance by implementing an automated system for fine-tuning SVM kernel parameters.

         Devised a comprehensive decision tree algorithm for Scottish Techmedia to deduce animal postures from tri-axial accelerometer data.

         Created an animal vocalization detection algorithm using Gaussian Mixture Models and Independent Component Analysis to effectively identify distress in animals.

         Successfully employed self-organizing maps and Hidden Markov models to infer animal behaviors.

 

 

Programming/Languages: MATLAB, Java, C++, R, SQL, HTML and JavaScript.

Data Mining: data preparation, analysis, cleaning, knowledge extraction, and interpretation.

AI/Machine Learning - Support Vector Machines (SVM), neural networks, decision trees, fuzzy logic, genetic  programming, self-organizing maps (SOM), and K-Nearest Neighbor.

Anomaly Detection: Support Vector Machines, Hidden Markov Modeling, Gaussian Mixture Models (GMM).

Time Series: Autoregressive and Moving Average (ARMA) modeling.

Statistical Methods: hypothesis testing and confidence intervals, sample size determination, linear and logistic regression, Bayesian networks.

 

 

Smiths Detection, Pasadena, California, 2007 to Present

Engineering Scientist

• Played a key role in development of innovative pattern recognition/machine learning algorithms for real-time toxic and industrial chemical detection.

 

Continued

Yyyyyy x. yyyyyy ~ Page 2 of 3

 

Professional Experience continued

 

• Resourcefully leverage machine learning techniques to achieve best detection rates and reduce false alarms from widely varying chemical sensors.
• Adeptly use signal processing techniques to measure, strengthen, and filter sensor signals.
• Conduct statistical analysis of sensor signal data to identify and recommend improvements to manufacturing of chemical sensors.
• Perform data fusion from sensor signals to monitor activities in remote locations for the U.S. Army; extensively utilize machine learning anomaly detection techniques on low-level sensor data streams to optimize activity and speech recognition.
• Exercise excellent communication skills to compile and present findings.
• Proficiently implement techniques spanning self-organizing maps, one-class SVM, decision trees and neural networks.

 

University of Strathclyde, Glasgow, Scotland, 2006 to 2007

Research Associate

• Proactively explored and developed machine learning/data mining systems for condition-based monitoring of livestock.
• Leveraged state-of-the-art sensor technology in support of multi-million dollar research effort to improve livestock productivity and reduce cost of farming.
• Retrieved, critically examined and organized data from databases using SQL.
• Effectively cooperated with clients to define and model animal behavior and develop appropriate activity and speech recognition systems.
• Prepared and presented progress report to project management team.

 

Durham University, Durham, County Durham, England, 2002 to 2004

Teaching Assistant

• Contributed to Computer Science Department by delivering course material to students.
• Primarily led class instruction for Programming Data Structures, Introduction to Java, and Information Systems.

 

Northern Precision Engineering, Newcastle, Tyne & Wear, England, 2000 to 2001

Systems Analyst

• Applied strong organizational skills toward implementing, testing and validating database solution for accurately monitoring material stock levels.

 

 

School of Computing & Technology, Sunderland University, England:

         PhD in Artificial Intelligence, 2006

         Master s in Computer Based Information Systems (Distinction), 2001

Key Areas of Study: Object Oriented Design, Software Construction, Project, Computing & Research Skills, Application Building, Software Environments, Systems Development, Network & Computer Hardware, Relational Database Management System, System Engineering.     

 

University of Nairobi, Kenya:

         BSc in Mechanical Engineering, 1997

 

Continued

Yyyyyy x. yyyyyy ~ Page 3 of 3

 

 

 

E.M. Mugambi & A. Hunter. (2006) Inducing Comprehensibility in Evolutionary Polynomial-Fuzzy Classification Models. 2^nd Symposium on Evolving Fuzzy Systems, 7-9 September 2006, Ambleside, Lake District, UK (draft paper accepted for presentation and publishing by IEEE).

 

E.M. Mugambi. (2006) Inference of comprehensible models in medical data mining. PhD Thesis, June 2006, Sunderland University, Sunderland, UK.

 

E.M. Mugambi, A. Hunter, L. Kennedy & G. Oatley. (2004) Polynomial-Fuzzy Decision Tree Structure For Classifying Medical Data. Knowledge Based Systems, vol 17, issue.2-4,pp. 81-87.

 

E.M. Mugambi, A. Hunter, (2003). Multi-objective Genetic Programming Optimization of Decision Trees for Classifying Medical Data.  KES2003, Oxford University, UK.

 

G. Oatley, J. MacIntyre, B. Ewart, E. Mugambi. (2002) SMART Software for Decision Makers KDD Experience. Knowledge-Based Systems 15 (2002) 323-333.

 

E.M. Mugambi. (2001) An Application of Neural Networks, Statistical Methods and Decision Trees in Data Mining. MSc Thesis, SCET C Sunderland University, UK.

 

 

Best Paper C AI 2003 Conference, Technical Stream, UK, 2001

 

 

Permanent US Resident