Project Name : RISK Analysis
Project Report : https://github.com/SJSU272LabS17/Project-Team-19/blob/master/Report/Team19_Project_Report.pdf
Web Application URL: ec2-54-187-49-103.us-west-2.compute.amazonaws.com
Driver Risk Assessment and Monitoring
The total number of automobile accidents and fatalities has been on the increase in most States. If a person has had many accidents in the past, car insurance companies and transformation network companies (like Uber, Lyft etc.,) logically assume that the person is likely to have more accidents in the future.
- The project aims to look at certain other data points along with the driving history and identify the risky and cautious drivers.
- It also facilitates monitoring the driving behavior and provide key signs of unsafe and aggressive driving practices.
Our solution aims to look at the below data points and identify the risky and cautious auto drivers
- DMV “points”: The point system used by DMV to monitor the driving habits of all licensed drivers.
- VIN (Vehicle Identification Number): VINs summarize information about the vehicle.
- Credit Score: Good credit indicates that one is responsible and pay bills on time.
- Location: City dwellers pay more for insurance because there are more people in cities, raising the likelihood of accidents.
- Age: Drivers under 25 years old and those over 65 years old are statistically more likely to be involved in accidents.
- Gender: As per stats men drive more miles on an average than their female counterparts, automatically raising the likelihood of collision.
- Marital Status: Married people are known to abide by the traffic rules and relatively safe than than the unmarried people.
- Driving experience: The more the experience, the safer the driver is considered.
- Annual mileage: More annual mileage, means more likelihood of being involved in accidents.
-
As an Insurance Company Personnel, I can assess the Risk category of the driver and also decide on the insurance premium to be charged based on this. I can also monitor his driving pattern on regular intervals. If proven safe and trustworthy, the insurance premium can be reduced.
-
As a Cab Service company Associate, I can check the past driving history of the driver before employing him and also monitor him regularly to assess his driving behavior. I can also check his driving mannerism in case of any complaint or concerns raised by my customers.
-
As an individual, I can also check my driving score, before applying to companies like Uber.
- System architecture for Risk analysis implememtation
- Driver behavior monitor architecture:
Node-RED workflow for driver behavior monitoring:
Map UI displaying vehicle trace and driving behavior segments:
Working Model on AWS
http://ec2-54-149-92-43.us-west-2.compute.amazonaws.com/
Team 19
=================================================
2) A mobile dashboard to control all connected IoT devices in the house
The number of IoT devices are increasing rapidly and it is becoming challenging to manage all these devices individually. To make our homes smart, the devices need to have a centralised control system. Our solution aims to build a dashboard or use voice controlled device like Watson, Alexa, Google Home to control and adjust different smart devices at home like lights, switches, music player, sprinklers. This solution will make our lives easier, more convenient and comfortable. We plan to implement the following features:
- Integrate the control of multiple devices to one platform.
- Create the dashboard.
- Enable voice controlled interaction with the smart devices.
Reference: http://ranjanr.blogspot.com/
- Knowledge Base for Small and medium scale Organization.(New Idea)
Thanks to Cloud IAAS, PAAS and SAAS Services available right now, many small and medium scale organizations are coming up in the market. But it is becoming hard for them to devote resources to organize the documents or knowledge base in the organization. It becomes hard for a new joinee to find the answers for all the small problems. By keeping this in mind we plan to implement a SAAS service for this kind of organization in the following way:
- Create an interactive Front-End where different people can log in to search for desired problem.(IBM SSO and LAMP Stack)
- Use IBM Retrieve and Rank to:
- Upload Documents
- Parse Document
- Upload Sample Question
- Train the Data to according to Sample Questions.
This SAAS application will be subscription based, and the user will be charged according the the number of questions asked. Though this will be useful for any industry type, below are few examples of industries which we can use use in our Application.
- Customer Care support
- Health Care
- Research and Development
- Support Team.
- Education
- Telecoms, and etc.