Mid-May marks the final project protection week for almost 200 students at FPT University Hanoi for Spring 2020. Here, various solutions for effect business operations had been successfully presented by the students.

1. Management system for small/family business

Most management tasks in small and household businesses, like management of orders, goods, and staff, are currently being done only via papers or on Excel. This method is extremely limiting in that it is cumbersome, with way-too-many documents to store, and time-consuming. Yet, applications of IT in these businesses to optimize management tasks are not quite popular.

While there are already some applications and software for this particular purpose, they have yet to solve the issues completely, or are just too expensive for popular usage. Acknowledging this, a team of students from FPT University including: Tran Quang Tien, Nguyen Vu Long, Nguyen Trung Son, and Nguyen The Hung, had built MSSFB (Management system for small/family business), which helps improve management efficiency, reduce involved costs and time, and is completely free.

Tran Quang Tien, Nguyen Vu Long, Nguyen Trung Son, and Nguyen The Hung presenting their final project at FPT University.

MSSFB utilizes TITYSS, and is developed for both Android and iOS – two most popular operating systems for mobile devices in the world currently. With the system, users can manage their inventory lists, view reports, and track received/delivered goods. They can also management employees by making staff accounts and delete these when needed.

TITYSS application interface before and after log in.

The team had also integrated quick searching and barcode scanning technologies on mobile, and comparisons with external barcode scanners had shown similar speeds in optimal conditions.

For analysis and reports, TITYSS allows daily revenue and costs analysis, as well as offers statistical reports of inventory and good values. It also lists products out of product, on sale, and those soon to be out-of-stock. Reports from the system will detail all information of goods out and in of stock, with clear amounts and prices.

TITYSS allows daily revenue and costs analysis.

A huge advantage of MSSFB is that it is extremely easy to use, with user-friendly website and application interfaces. However, its application requires devices to be using at least Android 5.0 and iOS 9.

When asked of future prospects, the team representative – The Hung, had shared: “In the near future, we will focus on designing the system with expandable databases, full business situation coverage, image processing integration, as well as AI and machine learning applications…”

MSSFB, with its benefits, will be promising for current small and household businesses. If successful in real-life application, the students’ project will be proven to be practical and satisfactory of market demands, rather than just “on paper”.

2. Intruder Detection System for business information security

According to a report by the Ministry of Information and Communications, in January 2020 alone, there have been 283 cyberattacks aimed at Vietnamese information systems. This had shown that cyberattacks aimed at devices are quite the pie for hackers. Their attacks are also getting more complex, sophisticated, utilizing various different methods, making detection, prevention, and countering of them harder than ever before.

Acknowledging this, a group of students including Hoang Hai Dang, Nguyen Minh Duc, Dang Vu Khoa, Nguyen Viet Anh, and Dau Anh Tuan from FPT University Hanoi, had together built an Intruder Detection System based on Adversarial Networks, using a Deep-learning model.

5 team members and their guidance teacher presenting the project.

The team had chosen to research the Generative Adversarial Networks (GANs) model, which is an algorithm structure using two Neural Networks. GANs are particularly popular in this generation of image, video, and voice. The model’s features make GANs extremely useful in detecting irregularities in network traffic.

All team members had discussed and selected their final model to be Iterative Incremental Process Model. While the project stops at only running the algorithm successfully, it had been highly acclaimed in both technical aspects and future prospects.

“Using GANs for IDS (Intruder Detection System), is extremely difficult, but not impossible. The reason is that the networks need to learn how to detect all known attacks, yet still use the builder to create new packs and trick the detection. Our project had used an advanced Deep-learning model that utilizes Generative Adversarial Networks, with good accuracy for detections of attacks. We believe that should the model be used to its full potential, it will be a strong IDS tool in network security. In the future, we further improve this technique, even using Deep Convolutional Generative Adversarial Networks (DCGAN) and apply them to IDS to see if it works better,” said the team’s representative.

While on the same topic of Information Security, the student team of Le Quang Sang, Do Hoang Long, Nguyen Quoc Hoang, and Nguyen Vuong Tuan chose to “Build a Data gathering System from Server Procedures and Supervision”, which aims to develop a safety network that catches incidental of intentional events caused by users’ carelessness. The target is to have a supervision network from servers, then a safe network for SMEs in Vietnam – those with stable financial platforms, or are just ignorant of cybersecurity, and thus are easy targets of cybercrimes.

Le Quang Sang, Do Hoang Long, Nguyen Quoc Hoang, and Nguyen Vuong Tuan after their presentation.

In the project, the team had used the free System Monitor service provided by Windows to recording running tasks on servers. They also wrote a tool that will send these data to the Servers and analysis and categorization of prevalent risks on those servers.

According them, while there are already solutions for these kinds of risks, they are mostly of foreign roots, and thus unfamiliar for most Vietnamese businesses. They aim to develop a system for Windows that can gather and supervise all activities from servers to detect and display irregularities to administrators – much similar to a part of EDR systems. And their targets are businesses that use Windows servers.

Talking about future plans for their product, the team leader, Le Quang Sang, had shared: “In the future, we will continue to develop the system, and transform it into a research article. But first, our thesis holds potentials for integration into FPT University’s examination system, for detecting usage of cheating software during tests. Then, we will move onto small businesses.”

FPT University’s graduation thesis protection is not only a big examination, but is also an opportunity for the school’s seniors to work in practical projects. It is also a golden time for juniors to discuss topics and gain experiences for their upcoming final year.

Minh Minh
FPT Education

Related posts: