96% of global cybercrime attacks can be attributed to ​phishing, ​yet there exists no viable solution to accurately fight phishing. These attacks cost the world $6 trillion annually, consequentially increasing brand erosion and reducing online channel adoption. We want to help enterprises counter phishing attacks to minimize financial losses and regain brand trust. Our solution is a detection algorithm that determines phishing activity. Through our unique use of ​computer vision​ to detect logos and related brand content, paired with HTML+URL classification, we provide clients with accurate and reliable detection scores. Doing so empowers them to secure their online presence.

It is no doubt that employees are the powerhouse of our companies. However, employee negligence and human error are some of the main causes of data breaches. This costs nearly $4 million on average for a business in the U.S., posing great security and financial risks. PrivaSee is a platform where security professionals can easily identify the exploitable employee public data and take action to mitigate the risks right away. This allows our clients to fully focus on their business objectives while they continue providing necessary resources for the employees.

With the rapid transformation to the digital age, one sector that has changed the least is the banking and financial sector. After the creating of the first cryptocurrency, there has been an eager push to use this technology to benefit humanity by creating new technologies that harness the power of a decentralized system. Every new technology is a double-edged sword. It can be used to benefit people as well as suppress them if used incorrectly. There have been efforts by organizations as well as countries to use cryptocurrency to collect personal data and evade sanctions.

Network traffic and accurate location data draw significant interest from network operators and many other industries for marketing and service improvement potentials. At the same time, such data draw significant scrutiny due to privacy concerns. This project attempts to address privacy concerns by building a machine learning model based on truly anonymized data where all privacy-related information is peeled off. The project solution helps improve consumer confidence by eliminating privacy exposure while allowing the sponsor to save significant resources by simplifying the data wrangling process from complex and disintegrated data sources.

Cube Rule is an AWS developer education program integrated with Security Innovation’s simulation platform, “CMD+CTRL”. Cloud technology has become extremely popular. But as technology becomes more advanced, so do hackers. Many developers don't learn about security until late in their careers, which leaves a wide opening for malicious actors. Cube Rule provides a gamified simulation for developers to look through a hacker's eyes and exploit AWS vulnerabilities. Developers of all skill-levels can utilize the program, as CMD+CTRL provides hints and a scoreboard. By teaching developers to defend their platforms, Cube Rule will create a safer future for everything online.

Today’s technical landscape is constantly changing; in many ways this increased innovation is beneficial. However, continuous innovation also means practices and technological norms are changing at a fast pace. The majority of today’s youth are growing up with technology firmly integrated into their lives. We see it as our duty to educate kids on these potentially dangerous norms and practices they are at exposed to on a daily basis. PrivacyPal serves as a modernized education tool to meet these goals: protecting and empowering kids to create an informed and healthy relationship with technology.

This application directly promotes the integrity of incident reporting metrics within the Department of Technology Services at Seattle Public Schools. Departmental resources such as providing professional training, supporting district-wide projects, balancing IT worker’s assignments, and more, rely heavily on these metrics to strategically and effectively execute within the organization. By providing an application separate from the currently misused incident system, these metrics will no longer be skewed. This ultimately affects classroom technology and its influence on student learning, where our project makes a positive impact for the teachers today as well as the students in current and future generations.

Consumers are buying IoT devices without understanding their security and privacy implications. Traditional mainstream reviews only showcase device features. Consumers remain uninformed about security, but The Smart Guide hopes to change this. Our solution is a buyer’s guide for smart home devices. It helps consumers make an informed decision based on feature, price, and most importantly security. By factoring in cybersecurity in the rating of a product we hope to drive consumers to purchase safer smart home devices in a smarter way.

Industrial Control Systems (ICS) are cyber-physical systems that allow operators to monitor and control complex industrial processes, such as those found in energy, manufacturing, and other critical infrastructures. ICS operators face numerous challenges, especially information overload, reliance on disparate information sources, and a lack of aggregated, consumable, and intuitive ways to gain insights on ICS vulnerabilities and risks to protect their assets. By leveraging thoughtful user-centric design, we worked with FireEye to create and organize a structure for meaningful relationships between vulnerabilities and assets, designing a user-friendly system to display information that is both easy to understand and actionable.

Our capstone project aims at applying advanced machine learning techniques to identify advertisement traffic from the cellular network traffic. The idea behind classifying advertising traffic is relevant in today’s business era where digital marketing has become the predominant medium to increase outreach and connect directly with customers. Our team developed a novel approach to label & predict this advertisement traffic (accuracy ~97%), thereby paving the way for further research into identifying and blocking malicious/spam advertisements. The relevant activities we performed included getting data from network captures, data pre-processing, feature extraction, ad-labeling, applying ML algorithms and evaluating model performance.