Jury Members:
Hamdi Dibeklioğlu
Shervin R. Arashloo
Innovation Expert:
Aras Bilgen
Abstract:
Agrobotix is a mobile app for
agronomists and farmers developed using
and Machine Learning framework in order
to detect diseases in grain and grape
plants and make crop maintenance
efficient. It includes recommendations
on pesticides and their quantities. This
app calculates the optimal time to
perform treatments by considering
weather conditions such as rain and
include reminders and records of all
crop operations for a particular farm.
Based on a location radius that includes
several neighboring farms and considers
these farms' conditions, the application
can notify users of possible exposure to
a disease or pest without explicitly
disclosing the source. Furthermore the
app offers a platform that connects
farmers to paid experts who provide
advice.
Group
Members:
Mustafa Bayraktar
Cihan Erkan
Zeynep Gözel
Simge Işıl Tabak
Abstract:
Argus Vision is an application for
security and statistical data
collection. Our main targets are
parents, company owners and bankers. Our
system consists of one video camera, an
Android/iOS mobile phone, and a computer
for desktop app. The users will be able
to see the camera’s view from the mobile
application and the computer. By using
the computer, user draws lines to
determine where s/he wants to get
notified when that line is crossed by a
person. To determine efficiency in work
places, lines can be drawn to the
company corridors to count employee
traffic during work hours. Parents with
little children can also use Argus
Vision by placing the camera to the
child’s room and drawing a line next to
his/her bed.
Recipient of the User Experience Award
Group
Members:
Efe Acer
Hikmet Demir
Mehmet Mert Duman
Talha Murathan Göktaş
Burak Yaşar
Abstract:
Autoshop is a photo-editing tool that
makes use of cutting edge deep learning
techniques to add objects to and remove
objects from images. The tool is
innovative in the sense that there is no
commercial product that can perform
context-aware addition and removal of
objects. The task of context-aware
object addition and removal is
challenging since the texture, opacity,
brightness and other photographical
details of the added object should match
those of the background image.
Currently, this process is handled by
professional photoshoppers and it is
quite time-consuming. Hence, Autoshop
will allow people to edit their photos
effortlessly.
Group
Members:
Elif Arslan
Hasan Doğan
Asena Tuğba Şahin
Ahmet Emre Zengin
Tunç Zerener
Abstract:
In today’s world, depression has become
an important problem and many people
have a negative perspective on life.
Even though there is a trend of
increasing unhappiness among people,
this case should not be the final
verdict on humanity. Besides
antidepressants, there should be other
scientifically backed up ways of making
people happier. As a group of Bee+, we
want to help the people. We'll show you
psychological techniques and coping
skills to overcome stress and anxiety,
and improve your mood. You can also use
our progress page to see how your
happiness hormones are affected by
activities, foods and events so you can
concentrate more on things that benefit,
and less on things that don't. To do our
bit, we're also making the Bee+ app free
for anyone who signs up.
Group
Members:
Fuad Ahmadov
Ismayil Mammadov
Shabnam Sadigova
Abstract:
Charin is a cross-platform application
that supports both IOS and Android
operating systems. You can use Charin to
socialize with people that seek helping
people in need. By using Charin, you can
create groups and events to that aims to
help people suffering from health
issues, natural disasters, or any sort
of charitable deed. Via Charin, you can
fund raise events that seeks to help
people who need it. Users can donate to
the fund raisers established by
authorized users. In addition, you can
support animal rights and protect
animals by using Charin by conducting
events to help them. Events and groups
can be formed based on the location of
the user. Other users can join groups or
mark if they are attending an event.
Based on the user’s location, nearby
events and groups. Also a user can be
able to search for specific events or
groups and filter them according to
specified criteria.
Recipient of the Social Impact Award
Group
Members:
Muhammed Salih Altun
Samet Demir
Ensar Kaya
Ahmet Malal
Faruk Şimşekli
Abstract:
Microsoft Word is used by millions
worldwide. Currently, Word can be
operated with mouse and keyboard
devices. To help users who cannot or do
not want to interact with keyboard and
mouse, e.g. disabled or visually
impaired, we have developed an extension
to Word, a Word Add-in, that allows the
users, in addition to typing with their
voice, to: insert new lines and
paragraphs; select, delete, copy, paste
and navigate inside the document; change
the text font; have the contents of the
text read to them; search for words
inside the document; search the web for
particular words; all by using voice
commands. We used Google’s
Speech-to-Text and Text-to-Speech APIs,
node.js and Word JS API in a
client-server architecture, along with a
context-free grammar to recognize
commands.
Group
Members:
Merve Kılıçarslan
Yağız Efe Mertol
Ege Özcan
Çağla Sözen
Murat Tüver
Abstract:
Coda is a mobile application designed to
make playing musical instruments more
sustainable by making instruments more
accessible in terms of cost and
mobility. Coda is a system that renders
a musical instrument and creates a
Virtual Reality environment on the
smartphone screen. When the mobile phone
is attached to a simple Google Cardboard
like headset without extra
functionalities, the user is immersed
into a virtual environment with a
virtual instrument. To imitate the
experience of playing an instrument
intuitively, user’s hands are tracked by
the smartphone’s camera and maps the
hand coordinates to virtual coordinates
to detect interactions with the
instrument. Different modes can be used
to compose custom pieces, play over
different songs to learn the song
through visual directives.
Group
Members:
Nursena Kurubaş
Mehmet Selim Özcan
Mehmet Sanisoğlu
İsmail Yavuzselim Taşçı
Ayça Begüm Taşçıoğlu
Abstract:
Csion aims to optimize people’s
decisions about a subject based on their
characteristic behaviors. Csion is using
Myers-Briggs Personality types test to
have an initial categorization of the
users' characters. However, personality
alone is not the only variable when a
person makes a decision. A detailed
analysis of the subject that the user
will decide on is also needed to provide
reliable suggestions. For this end,
Csion has two different part. First, a
chatbot part that uses NLP techniques to
give detailed analysis of a paragraph
that states the problem of user. Second,
a question-answer section for users who
do not wish to write a problem in detail
to get decisions about predetermined
problems by choosing a problem and
clicking checkboxes to answer several
yes-no questions.
Recipient of the Data Science Award
Group
Members:
Kuluhan Binici
Yusuf Dalva
Musab Gelişgen
Ebru Kerem
Zeynep Sonkaya
Abstract:
Driving is a fundamental activity of our
daily lives, yet it can be troublesome
due to uncomfortable situations faced
during driving. Although some vehicles
provide smart assistance features, the
majority of drivers do not yet have
access to such systems to avoid
problematic circumstances. DRIvision,
only requiring a mobile device, provides
a package of driving assistance
features. Particularly, it contains
traffic sign recognition, attention
tracking, lane tracking, and collision
warning features helping drivers to
avoid certain issues in traffic.
DRIvision provides necessary warnings
based on instant visual analysis as long
as the mobile device is appropriately
positioned. DRIvision is developed for
the Android platform to make certain
driving assistance features accessible
to millions of users.
Group
Members:
Zafer Çınar
Engin Deniz Kopan
Enes Varol
Enes Yıldırım
Jury Members:
Çiğdem Gündüz Demir
Hamdi Dibeklioğlu
Innovation Expert:
Mehmet Çakır
Abstract:
Faveo is an application that aims to
provide fun and educational time to
fully or partially visually challenged
people. Users can log in, solve quizzes
from different categories and even
upload their own quiz questions using
nothing but their voice commands. If the
user desires to use the on-screen
buttons with the assistance of Faveo
text-to-speech, they can do so easily
because of the big and consistently
placed buttons throughout the complete
user experience. As a team, we chose
this project because we wanted to help
the unfortunate people who do not have
the same opportunities in life as we do.
Currently there are more than 36 million
people who are visually challenged and
we would consider ourselves successful
if even one of them spends one better
day because of our application.
Group
Members:
Muhammed Safa Aşkın
Mehmet Mert Epsileli
Yusuf Samsum
Fırat Yıldız
Abstract:
FORSETI is a web application which is
used to detect fake news, photos,
videos, and tweets by using machine
learning and artificial intelligence
models, so it does not need manpower.
This means saving money. FORSETI is also
user-friendly. It can be used with 3
simple steps which are sign up, login
and analyze. Moreover, it has some
additional features such as source
verification. FORSETI can automatically
search the Internet for the sources of
texts and photos instead of manual
search, so users save time.
Recipient of the Sibel Özelçi Best Senior Design Project Award
Group
Members:
Sait Aktürk
Zafer Tan Çankırı
Erdem Ege Maraşlı
Doğa Oruç
Kaan Ünlü
Abstract:
Guido is a cross-platform mobile
application for indoor navigation. With
it, one can create floor plans/layouts
using AR technology, edit them live by
adding and subtracting new parts. On a
completed floor map, they can give one
of few 3D humanoid agents a skin they
wish and place it where they want on the
floor plan to fixed locations to serve
visitors with general information about
the entire map or some parts of it. The
3D agent is also used by visitors as a
guide of the establishment when they
wish to go somewhere, look for a certain
good, exhibition or item. In this case,
it acts in a nature similar to a GPS
navigator, leading the visitor where
they wish to go and upon arrival
informing them about their destination,
where such description exists.
Group
Members:
M. Ege Acıcan
Nurefşan Müsevitoğlu
Can Özgürel
Gökçe Sefa
Emre Sülün
Abstract:
Learn and Teach is an online tutoring
platform that aims to connect tutors and
students in an online classroom
environment. The online classroom helps
students to learn regardless of their
locations. Students can search tutors,
filter them by experience, reviews, or
fee. Students can schedule a tutoring
session according to availability of
tutors. Lectures are done in the online
classroom. The online classroom includes
collaborative whiteboard, collaborative
code editor, video chat, text chat and
screen sharing features. After
completing the lecture, students can
write reviews for tutors and this
affects the credibility of tutors.
Group
Members:
Mahmud Sami Aydın
Berke Egeli
Naisila Puka
Halil Şahiner
Abdullah Talayhan
Abstract:
LIBRA is a user-friendly,
platform-independent, and web-based
genetic filtering and annotation system
equipped with a genetic profile matching
platform, that can be quickly integrated
and easily used by medical institutions
in order to explore genetic variation,
detect and diagnose rare diseases, as
well as safely collect and store their
data. It consists of two main modules:
genetic annotation and filtering module,
and matchmaking module. By genetic
annotation and filtering module, users
can upload, annotate and filter VCF
files based on frequency, impact, and
scenario of variants. By matchmaking
module, the genetic information provided
by VCF and disease related phenotype
created by user is used to match
patients and notify the users according
to their preferred similarity level.
Recipient of the Innovation Award
Group
Members:
Zeynep Hande Arpakuş
Zeynep Ayça Çam
Muhammet Said Demir
Zeynep Nur Öztürk
Elif Beril Şaylı
Abstract:
LikedIt is a smart website for wathing
videos, which analyzes and interprets
facial expressions of viewers. It
estimates the viewers' liking level and
intensities of basic emotions while
watching videos, by analyzing facial
expressions in real-time. Furthermore,
for every session, LikedIt provides a
detailed feedback report showing how
much the viewer likes or dislikes the
corresponding video scene by scene,
which emotional expressions displayed in
each scene, and where the viewer focuses
spatially at each frame. Based on such
statistics, LikedIt is able to rank
videos in terms of viewers' appreciation
levels as well as finding the most liked
and disliked videos.
Group
Members:
Pınar Ayaz
Barış Can
Faruk Oruç
Ünsal Öztürk
Mert Soydinç
Jury Members:
Shervin R. Arashloo
Hamdi Dibeklioğlu
Innovation Expert:
Veysi İşler
Abstract:
This project aims to develop an
application which will semantically
reconstruct the geometry and the texture
of indoor areas (such as a living room)
in 3D and allow a user to display this
reconstruction either using VR or on a
conventional display device (a monitor
or a mobile phone display) using a video
scan of the area. The user will be able
to manipulate the reconstructed objects
in 3D space and will also be able to add
certain items to the reconstructed
scene.
Group
Members:
Kaan Altınay
Can Demirel
Arda Gültekin
Berk Mehmet Gürlek
Koray Gürses
Abstract:
Nature's Call is a mobile application
that aims to aid both beginners and
experts in camping. It offers both
online and offline features. Nature's
Call allows users to identify an unknown
animal that they have encountered in
nature by taking a photo from their
mobile device inside the application and
shows required information of the
animal. Moreover, it provides camping
tips and tutorials which some tutorials
are given in video format to provide
further help. Ultimately, Nature's Call
application encourages people about
camping, informs them to avoid camping
accidents, and facilitate their camping
experiences.
Group
Members:
Yasin Alptekin Ay
Ateş Bilgin
Cansu Canan Ceyhan
Berkay Karlık
Burak Erdem Varol
Abstract:
Verification of official documents is
one of the biggest problems of modern
life. NeedVerify is a web application
that provides reliable documents and
buy/sell processes in less time and
energy. It supports three kinds of
operations which are about certificates,
diplomas, and notary documents. The main
purpose of the system is to provide a
digital version of the notary
transactions and document verification.
These operations are handled by
blockchain technology and the support of
blockchain technology provides more
secure transactions as it is promised.
As a result, we recommend using the
NeedVerify system instead of
conventional notary systems.
Recipient of the Wide Impact Award
Group
Members:
Bartu Atabek
Utku Görkem Ertürk
Merve Sağyatanlar
Deniz Yüksel
Abstract:
We are human beings. And it is in our
very nature that we wonder, travel,
explore, and capture breathtaking
moments that form our memories. Although
we hope to remember these memories, they
are stuck in digital technologies called
clouds or drives nowadays. With
Neverwhere, we want to allow you to
share, and (re)live your memories. We
also present a new perspective of
re-living your memories. With
Neverwhere, your favorite moments live
on your iPhone, beautifully laid out. We
eliminate the boundaries between the
screen and the content to help you feel
more emotional. Advanced technologies
like ARKit, CoreML, and RealityKit give
you a new level of interaction and
experience. And with every moment you
capture you see more of this world.
Which all adds up to an unlimited
sandbox. Neverwhere. A new way of
(re)living your best moments through
time and space.
Group
Members:
Berk Ataç
Tarık Emin Kapla
Can Ozan Kaş
Gülnihal Muslu
Ali Kemal Özkan
Abstract:
Nomad is a mobile application which aims
to help travellers explore new places by
following a “for travellers, by
travellers” approach, in which people
can create routes, and others can search
a route according to their preference of
what type of journey they would like to
go on and where they want to explore.
Users can add stops to the routes they
create on specific locations that they
seem worthy of attention, which act as
“checkpoints” for other users who follow
that route. Using augmented reality,
users that follow routes can also
collect tokens with using their cameras,
those who obtain such tokens can get
achievements.
Recipient of the Productivity Award
Group
Members:
Ahmet Abdullah Avcı
Faruk Ege Hatırnaz
Hygerta Imeri
Sıla İnci
Elif Gülşah Kaşdoğan
Abstract:
We come up with many new ideas every day
but only a few of them survive in the
long term. We think about the things we
want to do but forget them long before
we realize the opportunities around us.
NOTIVE is a mobile app that keeps track
of different types of activities you
would like to do and reminds you of them
through motivational notifications
whenever there is an opportunity to
accomplish your wishes. What makes
NOTIVE special is that it combines your
'normal' to-do list with a map and
motivational notifications to provide a
unique solution that makes your life
easier and helps you to manage your free
time, your trips, your wishes in a more
enjoyable way. So, don't miss
opportunities, they are actually really
close…
Group
Members:
Ömer Faruk Babademez
İdil Hanhan
Beyza Kalkanlı
Berfin Küçük
Kerem Yılmaz
Abstract:
Photonom is a mobile application that
provides several tools allowing users to
modify their photographs interactively
and in one place whilst conserving the
true nature of them. Users can take
photographs using the application or
upload them from the gallery. Using
Photonom, unwanted objects or regions in
photographs can be removed, effects of
rain and haze can be eliminated,
multiple images can be stitched
together, displayed facial expressions
can be modified, and the style of an
image can be transferred to other
photographs. During editing, users can
undo the applied operations. In
addition, quality of photographs can be
evaluated using Photonom.
Group
Members:
Gökberk Aktulay
Emre Keskin
Yağmur Özkök
Eren Yalçın
Abstract:
Our application is mainly a car-pooling
application so that students can travel
together, which will be both available
in GoogleStore and AppStore in near
future. Therefore, our application is
very similar to already existing
carpooling apps such as Uber and
BlaBlaCar. With this application
students who use buses, will no longer
need to get up really early to go to
ring stop then go to school with ring.
Instead they will meet with other
students who own a car and travel with
them. Some distinctive features of our
application are it is uniquely created
for Bilkent University students and it
suggests the optimal rendezvous point so
that trip is as short as possible.
Group
Members:
Huseyn Allahyarov
Cenk Er
Nurlan Farzaliyev
Ibrahim Mammadov
Mahammad Shirinov
Abstract:
Rendt is a platform for allowing remote
job execution in exchange for a fee.
There are two kinds of users - leasers
who offer their computational resources
for rent in exchange for an hourly rate
that they specify, and renters who are
willing to pay the fee and user leaser’s
computers to run their jobs on. The job
files get transferred through our server
and get executed in a safe container on
leaser’s machine, and the outputs of the
job are again uploaded on our sever. The
renter then can access their output
files after they’ve completed the
required payment.
Group
Members:
Masna Ahmed
Muhammad Umair Ahmed
Tunar Mahmudov
Dawood Muzammil Malik
Talha Zeeshan
Abstract:
SaySL is an easy-to-use smartphone
application that attempts to bridge the
communication gap between the hearing
and the Deaf communities. SaySL allows
people from the Deaf community to be
understood by non-sign-language speakers
by translating American Sign Language
gestures into the closest equivalent
English text. Unlike the current
competition in the market requiring
special equipment that is almost always
both expensive and heavy, our solution
needs only a camera-equipped Android
smartphone to accomplish the said task!
Group
Members:
Fazilet Simge Er
Giray Baha Kezer
Kaan Atakan Öztürk
Melih Ünsal
Abstract:
SoFarmSoGood is a farm friendly software
that intends to encourage farmers to
produce and sell their harvest. In
Turkey, sugar beet producers face with
quota problem.Encouraging local farmers
of a harvest into whole purchasing
system is an undeniably important issue
for agriculture supported economy.We
offer secure, robust and scalable
blockchain system, which is named
SoFarmSoGood.In our solution, we have
three main actors; farmer, cooperative,
company. Through virtual cooperatives we
offer an innovative solution to local
farmers for passing the quota
limitation. Thus, more local farmers can
become part of production chain.We
always need to remember, having happier
and more welfare farmers enables happy
agriculture and having happy agriculture
enables national welfare and happiness.
Group
Members:
Fırat Ege Akın
Fatih Çelik
Alp Ertürk
Burak Korkmaz
Abstract:
The aim of SolarUpp is to create a
platform where real estate owners can
get an idea of how suitable their
residency is for solar energy, and solar
system owners can track the performance
of their systems compared to our
estimates. We tried to make our solar
feasibility estimations highly accurate
while maintaining a pleasant user
experience. In particular, we used image
processing technologies to calculate the
available space in the roofs for this
purpose.
Group
Members:
Betim Doğan
Sera Fırıncıoğlu
Berfu Deniz Kara
Mert Özerdem
Alara Yaman
Abstract:
StudyB is a mobile application that aims
to create a student-based study platform
where specific courses are offered and
students are able to chat with each
other by entering one of the available
course chatrooms. StudyB is not only a
collaborative platform, but it is also
safe. More specifically, it is only open
to Bilkent University students to sign
up. Currently, Computer Science major
courses are available only but we expect
this to be widespread throughout the
campus. Users are allowed to enroll in
any course they prefer. Furthermore, our
application is designed in a
user-friendly manner to be simple and
understandable for everyone. Our
application is mostly beneficial for
those who need to catch up with their
classes, or do not have any friends to
discuss in case if they need help in a
specific homework or subject.
Group
Members:
Bilal Siraj
Kasymbek Tashbaev
Aliyu Saifullah Vandana
Abstract:
Sum of Sounds is a text to speech
conversion tool for mathematical
formulas and equations. Intended for
those with visual impairment, it makes
text to speech synthesis of scientific
documents a reliable option with
Google's speech synthesis technology at
its core. The software leverages the
power of LaTeX, a typesetting system
that is the standard for the
communication and publication of
scientific documents. Having LaTeX as
the underlying language makes it
possible to predictably and accurately
convert an image or document containing
formulas to a readable script, similar
to a transcript of a person reading them
out loud. Finally, an audio file of the
script being read in a voice and pace of
their choice is delivered to the user to
listen to.
Group
Members:
Burak Erkılıç
Mert Sezer
Okan Şen
Berk Yıldız
Abstract:
Writing code or reviewing one can be
quite tedious and energy-draining at
times and this tiresome process can
result in revisions or codes to be
faulty and even inefficient. The time
these actions require grows
exponentially into tremendous amounts as
the complexity of the work increases. In
order to soften the burden of these
works, The Game comes up with several
innovative ideas. The Game focuses on
creating a more appealing environment to
learn about reviewing codes using game
modes such as; challenges within classes
or in a single form, defect detection.
Teachers will create and post new
challenges thorough pathways and codes
to be reviewed, while the students will
have to work their way around these
challenges and find the defects in these
faulty codes.
Group
Members:
Ahmet Ayrancıoğlu
Deniz Dalkılıç
Kaan Gönç
Abstract:
The main goal of Newspector is to
provide people with a simple and
intuitive way to follow the latest and
the most important news from all around
the world. Newspector retrieves the
latest breaking news from the twitter
profiles of various verified news
sources. After that, Newspector analysis
the sentiments and the categories of
these news headlines, and clusters the
relevant news headlines together using
state of the art machine learning and
natural language processing algorithms
and tools. In addition to this analysis,
Newspector also keeps statistics and
metadata about news sources and the news
headlines they share. These statistics
are then displayed together with the
created newsgroups to our users for them
to compare both news sources and news
stories with each other.
Group
Members:
Mertkan Akkuş
Ahmet Furkan Bıyık
Ramazan Mert Çınar
Yaman Yağız Taşbağ
Emin Bahadır Tülüce
Abstract:
Thoracle is a tool to find tweets from
their screenshots. The tweets can be
searched through web application, mobile
application or chrome extension. After
you find the tweet, you can browse it on
Twitter. Thoracle also provides
different kinds of information about the
tweet: The author of the tweet is
classified as a bot or genuine account,
the relevant news articles about the
tweet content are listed, and there is a
comment section for discussing the
validity of the tweet’s content. The
main goal of the Thoracle project is
preventing public manipulation by fake
news through social media. We hope to
achieve this goal by providing the
fastest way for users from tweet
screenshots to the actual tweets along
with the analyses.
Group
Members:
Mert Çerçiler
Yiğit Kutay Gülben
Doğacan Kaynak
Burak Kırımlı
Servan Tanaman
Abstract:
The increasing price of the food and
beverage industry has a huge effect on
peoples' daily lives. When it comes to
socializing, the economic side of having
fun sometimes becomes a burden on one's
shoulders. Therefore, restaurants have
special campaigns to attract their
potential clients to visit their
restaurants. They put their low priced
or under-sold products to the campaigns
for a specific time interval. So,
Toproffer came as an idea to help people
to save their time and preserve their
economy with less effort compared to
searching restaurants blindly on the
streets. In a sense, Toproffer will
similarly bring the streets’ view as
Google Maps does to the users, in
addition to that it will act as
assistance for finding promotions
related to restaurants.