Deep Learning For Business
SHELDON AI wants to make deep learning a practical technology for businesses by making the implementation of artificial intelligence extremely easy for them.
Projects
MORTGAGE INDUSTRY
Correspondent Lender
Sheldon AI analyzed a large, industry-leading mortgage broker’s Loan Origination Software (Calyx Point and Ellie Mae Encompass) to identify trends using age, income, and similar industry positions. These inputs allowed us to predict increases in consumer income, create customized marketing strategies, and predict which existing customers could potentially become upgraded buyers. These consumer insights, in addition to research and development on new loan products and improvement in tracking milestones during the customer loan process, ultimately resulted in stronger leads, increased sales, and more profit.
In addition to our focus on deeper and predictive consumer insights, our analysis also helped identify potential data anomalies and worked to mitigate risk with FINRA. With our ability to assess both historical and current loan data we worked to make early predictions of any FINRA compliance concerns to mitigate potentially massive fines or licensing issues.
Service Industry
Home-Sharing for Professionals
Serving largely travelling nurses, Denver-based Furnished Finder wanted to identify both individual customer and industry trends in relation to popular locations and high-demand localities for their clients. Utilizing historical data, Sheldon AI wanted to predict where a nurse may travel next based on client preferences on hospitals, type of work, climate, geographic location, etc. This information would be tremendously valuable to medical staffing providers. Our data analysis determined several data-specific recommendations to Furnished Finder that would allow their data to be more robust, accurate, and AI-ready. This resulted in more stakeholder awareness and “buy-in” toward data solutions which Furnished Finder found significantly valuable to the growth of their organization.
Financial Industry
Private Equity Firm
Sheldon AI effectively analyzed long-term equity tick data to predict price movement of key industry leaders and determine trends of other equities within the same industry. This was accomplished by utilizing the coding language Python to develop trading platform API integration and Natural Language Processing to determine market sentiment and financial records analyses. Our proprietary deep learning/AI technology has proven valuable in both long/short strategies and option strategies. For example, Sheldon AI achieved 81% accuracy when using specific currency studies in correlation with the AUD and the USD (Australian dollar vs. US dollar).
HEALTHCARE INDUSTRY
Medical Technology Company
Through an open-information forum, Sheldon AI identified nerve structures from neck ultrasound images for Halyard Health, a large healthcare provider. With labelled ultrasound images, our proprietary neural network structure identified nerves which can be targeted for inserting pain management catheters. The reason for this process is to avoid anesthesia or strong pain-relief medicines for medical procedures by identifying the nerve from the ultrasound image and using pain management catheters as an effective alternative. This case study highlighted the Computer Vision image recognition ability of Sheldon AI.
MANUFACTURING INDUSTRY
Bakery Product Manufacturer
Through an open-information forum, Sheldon AI forecasted inventory demand on a weekly basis at multiple store locations for the multinational bakery product manufacturing company Grupo Bimbo. This process was critical to minimize the waste of perishable products. The goal of this project was to predict how much of each product will be needed for next week’s delivery. This predictive analysis would greatly assist store personnel in ordering the correct quantity of multiple items and ultimately reduce cost and increase profit. Predicting the need for raw materials on a consistent and accurate basis would also establish the platform for a fully-automated machine-learning-driven delivery method.
INSURANCE INDUSTRY
Major Insurance Company
Through an open-information forum, this case study focused on the identification of distracted drivers for State Farm. Utilizing our Computer Vision image recognition strategy, Sheldon AI took data from individual vehicle cameras to discern whether a driver was distracted and what the cause of the distraction may be. The goal of this project was to develop a system which can alert drivers when it finds the driver is not paying attention, enabling more focused, safer driving and, ultimately, reducing the number of annual deaths due to negligence while driving.
DENTAL INDUSTRY
Dental Care Provider
Through Sheldon AI analysis of data from leading dental practice management software Dentrix, we were able to provide a dental clinic with a deeper understanding of their customers – their schedules and their desired payment terms – and more accurate estimates toward insurance payouts. This was done by analyzing the following:
- Assessment of customer demographic including age, location, gender, and insurance status
- Breakdown of customer payments based on insurance status
- Claims assessment and considerations regarding which insurance companies are most likely to pay and in what time frame
- Assessment of the quantity of various procedures and the likely timing of completion
- Amount of procedures completed per dentist and total charges
- Assessment of customer appointment requests
- Calendar assessment and monthly appointment trends, including a month-to-month assessment of payments and insurance claims processed
- How many appointments will be made?
- How many appointments will be missed or cancelled
- Which procedures will be done?
- Which claims will be paid to maximize revenue?
- Which patient will pay a higher percentage of their bill themselves?
- Procedure recommendations based on Predictive Analytics performed on data
- Probability of the customer making payment in a defined timeframe
- Probability of customer payment once account is in collections and percent of accounts receivable recovery
SHELDON AI
Proprietary data platform
Our proprietary data platform distinguishes us from other data science firms by allowing us to complete projects with more agility and punctuality.