Position: Expert Data Scientist Specialization: Unsupervised Fraud Detection/Outlier Detection Type: Part-time that may become full-time(minimum 20 hours/week) Please find here a blog that is an introductory to the challenge that we're handling with: https://medium.com/dataman-in-ai/2-features-for-healthcare-fraud-waste-and-abuse-7c262ac59859 ** The project has already commenced. The data scientist must be prepared to join our team and possess skills in rapid learning, agility, and experience.
Company Overview: Our company is a cutting-edge insurtech-healthtech company that is revolutionizing the way of finding frauds and diagnosis. We are dedicated to leveraging advanced machine learning, deep learning, and artificial intelligence techniques to solve complex problems in insurtech-healthtech. As we continue to grow, we are seeking a talented and experienced ML/DL/AI Expert Data Scientist to join our team and play a pivotal role in enhancing our fraud detection and outlier detection capabilities.
An example of this project can be - If we have 3 cases where Chest X-RAY (medical test) was ordered - in the first, patient has infection and fever and Chest X-RAY was ordered. In the second case, a procedure was scheduled to a patient and Chest X-RAY was ordered. In the third case, a patient came to the hospital with a broken leg, and Chest X-RAY was ordered. The third case is an outlier, because a broken leg and Chest X-RAY are not related. We need to flag on the third case, and specifically on the Chest X-RAY.
About the data - structured dataset consists majorly Boolean features of medical conditions, medical history, medical test, demographics, financial.
Role and Responsibilities: As an ML/DL/AI Expert Data Scientist specializing in Fraud Detection/Outlier Detection, you will be responsible for:
Model Development: Design, develop, and implement models, etc. for fraud detection and outlier detection.
Feature Engineering: Wise Feature Engineering to add to our working pipeline, we would like to add a couple of additional features, add additional model that will cover some other angels, for example, similarity model (Spectral/Contextual Clustering), and time series analysis model, and plug them into our current processes.
Learning Event Sequences: Utilize your expertise in handling learning event sequences and similarities to build models that can effectively capture and analyze sequential data patterns for fraud detection.
Collaboration: Collaborate with cross-functional teams including data engineers, domain experts, and software developers to integrate the developed models into our existing systems and applications. Contribute to a cohesive team environment that fosters knowledge sharing and innovation.
** It is not anomaly detection, it is fraud detection.
Qualifications:
- Proven experience (at least 3 years) working on fraud detection and/or outlier detection projects, Python and Data Science. - Strong background in Feature Engineering, handling imbalanced datasets and designing solutions to address data class imbalance. - Proficiency in programming languages such as Python, ML - Experience in analyzing and interpreting complex learning event sequences for fraud detection purposes. - Exceptional ability to communicate complex technical concepts to both technical and non-technical stakeholders. - Prior experience with insurance and/or healthcare data is a significant advantage. - Strong problem-solving skills and the ability to work independently in a dynamic environment.
Benefits:
Opportunity to work on cutting-edge projects that have a meaningful impact on healthcare. Collaborative and innovative work culture that values learning and growth. Access to the latest tools, resources, and research in the field of ML/DL/AI. If you're a passionate and experienced ML/DL/AI Expert with a focus and experience with Fraud Detection/Outlier Detection and are excited about contributing to the growth and success of our Company, we encourage you to apply. Join us in our mission to revolutionize healthcare through advanced data science and AI techniques.