School of Computing Student Conference 2018

The 2018 School of Computing Student Conference was a great success – students from across the School showed their final year projects, their PhD presentations and their coursework to an audience of university and business guests. This was one of the first events in the new Future Technologies Centre, and thanks to Mike Reid (Information Services) we were able to make full use of the new facilities.

Prizes were awarded for the best work, judged by SoC staff:

Best software demonstration:  Douglas Farr: Applying Block Programming to SQL
Honourable mention: Kees Jansen: A visual representation of our universe (Virtual Reality project)

Best poster: Jason Bissessur: Artificial intelligence in business, effects on employment
Honourable mentions:
Delga Denies: How secure are we under the small printed terms and conditions?
Jackson Jacob: Progress Web Applications for the student housing market

Best presentation: Alaa Mohasseb: A Customised Grammar Framework for Text Classification
Honourable mention: Xudong Wang: Mastering Gomoku from Scratch by Applying AlphaZero Algorithm and Generalised Representation

Abstracts for all the presenters are to be found on this website.

 

 

Posted in SC2018 | Comments Off on School of Computing Student Conference 2018

Cloudy with a Chance of Drones

Stuart Hotchkiss
BSc Forensic Computing

During recent years Unmanned Aerial Vehicle (UAV) usage has grown in military and consumer fields. Whilst we hope that good security practices are being implemented into consumer drones, it seems that this may not be the case, giving the chance for attackers to exploit flaws in consumer drones. This project investigates Drones controlled via WI-FI that have a video feed used for the control of the UAV, as these seem to have a great susceptibility to hijack, jamming and spoofing attacks. It is assumed the video feeds are not protected by any type of encryption and are therefore vulnerable to hacking and/or spoofing. The project also explores whether owners of drones know of these flaws.

As  more and more businesses are investing use of Drones, this project highlights the impacts of the flaws to the businesses if these exploits are not mitigated. A brief pen test is conducted on a targeted drone that is widely used and recommended in the “Drone Community” giving a severity rating, and brief mitigation strategies are suggested for the findings.

.

Posted in SC2018 | Tagged , , , , , , | Comments Off on Cloudy with a Chance of Drones

Developing​ ​a​ ​therapeutic​ ​companion​ ​with​ ​the​ ​use​ ​of​ ​artificial​ ​intelligence,​ ​to​ ​improve cognitive​ ​and communication​ ​skills​ ​in​ ​children

Yasmeen Hazy
BSc Software Engineering

An Alexa Skill that reads stories with children before they are at the age at which they can read themselves, asking the children questions about the story as they go along to allow learning of letters and words, as well as understanding of the storyline.

Posted in SC2018 | Tagged , , , | Comments Off on Developing​ ​a​ ​therapeutic​ ​companion​ ​with​ ​the​ ​use​ ​of​ ​artificial​ ​intelligence,​ ​to​ ​improve cognitive​ ​and communication​ ​skills​ ​in​ ​children

Designing a Tailorable Email Analysis Tool

Harry Senior
BSc Forensic Computing

The purpose of this project was to create a tool that can be used to analyse large quantities of emails for useful statistics. Most if not all companies have some sort of email policy requiring employees to be punctual with replies, conscious of possible malicious emails, and to keep email usage professional and appropriate. Enforcing this can be difficult, requiring HR to sift through many emails in order to identify issues. As the uses for a tool like this are broad the goal was to create a platform to allow a user to modify analysis methods to accomplish their goals and identify trends. Two versions where programmed, the second being an improvement on the first, and a third version was designed to be an improvement on the second. The program was designed to allow direct modification of the database, the methods used to insert data, and the procedures used to generate statistics. Creating a versatile platform for analysis means each deployment of the program can be modified to accomplish different goals.

Posted in SC2018 | Tagged , , , , , , , , , | Comments Off on Designing a Tailorable Email Analysis Tool

Polarity/emotion detection in social media data

Kyle Minter,
BSc Forensic Computing

Emotions are part and parcel of human life and highly influence decision making.  Computers have been used for decision making for quite some time now but have traditionally relied on factual information.  In this project, a machine learning approach was used to recognize 10 basic emotions (Empty, Fun, Happiness, Hate, Love, Neutral, relief, sadness, surprise and worry) using a heterogeneous emotion-annotated dataset which was collected from twitter.  From this results were collected and analysed to show the effects of punctuation and numbers. Further analysis showed how the program has assigned the emotions to the text and where it has made mistakes.

 

Posted in SC2018 | Tagged , , , | Comments Off on Polarity/emotion detection in social media data

StarBet: Betting and authentication app

Lance Biddall
BSc Computer Science

Betting has become more popular in the digital world with the release of smartphones, ease and accessibility of applications and the freedom of on the move digital interactions. Now it is becoming less common for people to physically go into a betting establishment to place a bet. With new successful apps being created at an exponential rate, there is the potential to creating a unique betting app which is used solely for the users, removing 3rd parties and their financial incentives.

This project application called “Starbet” aim to firstly Create a betting game that is used within a social setting, where users can bet and authenticate between themselves to win game currency and not real tender. The final aim is to give the users an understanding of betting and the potential drawbacks from a protected environment. “Starbet” application uses Firebase real-time database for instant database manipulations to enable users of android devices to rely on instant updates to bets.

Posted in SC2018 | Tagged , , , , | Comments Off on StarBet: Betting and authentication app

Text classification for detection of potential threat

Fatima Chiroma
PhD student

The social network has become a vital medium for communication. People have the freedom to share their opinions using different online social network platforms. This has however led to some growing concerns regarding its negative impact on human safety, for example cyberbullying and the spread of suicidal ideation.

Therefore this work focuses on the application of machine classifiers: Support Vector Machine (SVM), Naive Bayes (NB), Decision Tree (DT) and Random Forest (RF), on suicide-related tweets, and their accuracy in identifying and distinguishing suicidal communications. The results of the experiment show that SVM has the best score for Precision, while NB has the best Recall score, F-measure and Accuracy. Overall, NB has slightly outperformed the other classifiers in three of the four standard classification measures: Recall, F-measure and Accuracy.

Posted in SC2018 | Tagged , , | Comments Off on Text classification for detection of potential threat

Advanced Analysis of Darknet Forums

Tori Tompkins
BSc Forensic Computing

The darknet, or darkweb, is a subset of the deep web made up of several networks (eg Tor, Freenet, I2P) which encrypt, decentralise and anonymise network communication specifically for illicit purposes. Forums consist of around 5% of the Tor darknet yet are largely unexplored in current academia.

For the purpose of this paper, a python web crawler was created to collect the heterogeneous communication data from three different forums on the darknet, Tor. The 60,000+ threads and posts, stored in an SQL database, have then been analysed using three major data science concepts: social network analysis, latent semantic analysis and statistical analysis to gather an insight into how and why darknet forums are being used.

Posted in SC2018 | Tagged , , , | Comments Off on Advanced Analysis of Darknet Forums

Gem-Checker

Jamie Cleare
BSc Computer Science

Many programming languages today use package libraries that provide developers with a
number of easy to use, and implement, features and functionalities. However not maintaining these libraries can lead to the introduction of vulnerabilities and deprecated code in software applications, as well as causing a variety of issues when upgrading software to the latest versions.
This project proposes a notification alert package for the purpose of identifying out-of-date
and poorly maintained package libraries implemented on Ruby on Rails applications,
otherwise known as Gems. The package developed in this project will analysis the current
versions of Gems installed on Ruby on Rails applications and compare them with the most
recent versions available from the ruby gem hosting service, RubyGems.org. Once
analyzed, the package will produce a variety of notifications, alerting developers to the
version status of Gems installed on their applications. This project hopes to support
developers with maintaining library packages used within their applications and help gain a better understanding of the Gem packages themselves.

Posted in SC2018 | Tagged , , , , | Comments Off on Gem-Checker

Progress Web Applications for the Student Housing Market

Jackson Jacob
BSc Computer Science

UCASA aims to unify the student housing experience into a single platform on a national scale,
targeting three distinct demographics, Students, Landlords and Universities. Its aims include
helping students in making more informed decisions on properties, streamlining property
management so the landlords can get on with the sweeter things in life and encourage Universities
to adopt a preventative role instead of a reactive one within the student housing market.
For UCASA to achieve this, the project will utilise the latest in Web Technologies to develop a
product that is available on all platforms, be it a mobile, tablet or desktop devices while adopting
the most recent design standards for a seamless experience from day one.
Posted in SC2018 | Tagged , , , , , , | Comments Off on Progress Web Applications for the Student Housing Market

Speech To Text for the hard of hearing in Higher Education

Gregory Harrison
BSc Computing

In higher education there is an accessibility issue that exists for students who are hard-of-hearing. We proposed a method of manipulating an existing open source API to create an accessibility tool that will aid these students to still take part in lectures. We looked at existing research in speech recognition and discussed the methods, benefits and limitations of the current technology and looked at applications that exist already which are aimed towards assisting users in note-taking or ease-of-use. The development of our own system which aims to bridge this accessibility gap was detailed. After this, we then analysed the performance of our developed system and discuss its use and limitations.

Posted in SC2018 | Tagged | Comments Off on Speech To Text for the hard of hearing in Higher Education