Theses and Projects

From NET Wiki
Jump to navigation Jump to search

An introduction to the Computer Networks group

See a poster for a general overview, an anchor to our research activities, a list of social computing related or networking-related publications, and the annual report(s) for our recent activities.

Open Theses and Student Project Topics

The Computer Networks Group is always looking for motivated students to work on various topics. If you are interested in any of the projects below, or if you have other ideas and are willing to work with us, please don't hesitate to contact us.

  • (B) Bachelor thesis
  • (M) Master thesis
  • (P) Student project


* New! 3D natural hazard simulator

The aim of the project is to simulate representative natural hazards for hazard response, such as flooding and forest fire. A natural hazard response simulator will be implemented for both visualization and performance validation. For example, we can visualize the flooding of 2021 in Germany, and then validate the performance of drone deployment in hazard sensing and emergency communication. Here, we introduce some related works in virtual 3D scene which may help you to understand this project, e.g., Agents Toolkit (ML-Agents) [1], DisasterSim [2] and Airsim [3].

[1] Unity Technologies.Unity ML-Agents Toolkit. Jan 26, 2021.URL:https://github.com/Unity-Technologies/ml-agents. (accessed: 21.11.2021) [2] Wang, H., Liu, C. H., Dai, Z., Tang, J., & Wang, G. (2021, August). Energy-efficient 3D vehicular crowdsourcing for disaster response by distributed deep reinforcement learning. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining (pp. 3679-3687). [3] S. Shah, D. Dey, C. Lovett, and A. Kapoor. “Airsim: High-fidelity visual and physical simulation forautonomous vehicles”. In:Field and Service Robotics. Springer. 2018, pp. 621–635.

Please contact Prof. Xiaoming Fu [fu@cs.uni-goettingen.de](B/M/P) Dr. Tingting Yuan [tingting.yuan@cs.uni-goettingen.de] (B/M/P)

* New! Video analytics with deep reinforcement learning

The proliferation of video analytics is facilitated by the advances of deep learning and the low prices of high-resolution network-connected cameras. However, the accuracy improvement from deep learning is at the high computational cost. Although the state-of-the-art smart cameras can support deep learning method, the deployed surveillance and traffic camera paint a much bleaker resource picture. For example, DNNCam that ships with a high-end embedded NVIDIA TX2 GPU costs more than $2000 while the price of deployed traffic cameras today ranges $40-$200; these cameras typically loaded with a single-core CPU only provide very scarce compute resource. Because of this huge gap, typical video analytics applications, e.g., traffic cameras stream live video to remote server for further analysis. As a result, a natural question occurs: which video streaming configuration also server decoding configuration should we select to guarantee high analysis accuracy as well as not incur network congestion? To answer this question, we attempt to explore the performance of deep reinforcement learning under this scenario.

Please contact Dr. Tingting Yuan [tingting.yuan@cs.uni-goettingen.de], Weijun Wang [weijun.wang@informatik.uni-goettingen.de] (B/M/P)

* New! Convolutional neural networks and transfer learning for change estimation with satellite images

Satellite images are a popular input to estimate wealth measures (e.g. income or consumption) on the spatial scale, so to determine which locations are richer or poorer than others within a certain time interval. However, the use of these images for estimation of the changes in these measures over time for given locations is investigated only insufficiently. This project/thesis topic intends to address this problem by applying Convolutional Neural Networks (CNNs) with Transfer Learning on a small data set of images from villages in Thailand and Vietnam. Among other things, this topic contains experimental comparisons of different approaches and CNNs in both Regression and Classification.

Please contact Fabian Wölk [fabian.woelk@cs.uni-goettingen.de](B/M/P)

* New! AI for Games

Can Artificial intelligence (AI) beat humans at games? AI has played an increasingly prominent and productive role in the gaming world. Implemented in many different ways, AI is used to improve game behaviors and environments. In this project, we will design AI algorithms (i.e., multi-agent reinforcement learning) for games (e.g., StarCraft: https://github.com/oxwhirl/smac). The main challenge here is to coordinate agents in achieving joint goals (i.e., win), such as by efficient communication.

[1]https://www.nature.com/articles/d41586-019-03298-6

Please contact Dr. Tingting Yuan [tingting.yuan@cs.uni-goettingen.de] (B/M/P)

* New! Socioecomonic analysis based on spatiotemporal and linguistic analysis on microblogging data

Identifying the socioeconomic status (SES) of users in social media like Twitter or Weibo is useful e.g., for digitized advertisements and social policies. This study aims to collect profiles of Twitter users on selected topics such as culture or foreign language learning, extract the temporal, spatial and linguistic features, and compare different classification algorithms (e.g., decision tree, random forest, na\"{i}ve Bayes, deep learning, and Gaussian processes classifier) to predict the socioeconomic status.

[1] Ren Y, Xia T, Li Y, Chen X. Predicting socio-economic levels of urban regions via offline and online indicators. PLoS One. 2019;14(7):e0219058. Published 2019 Jul 10. doi:10.1371/journal.pone.0219058 [2] Pappalardo L, Pedreschi D, Smoreda Z, Giannotti F. Using Big Data to study the link between human mobility and socio-economic development. In: IEEE International Conference on Big Data 2015. doi:10.1109/BigData.2015.7363835 [3] Vasileios Lampos, Nikolaos Aletras, Jens K. Geyti, Bin Zou and Ingemar J. Cox (2016). Inferring the Socioeconomic Status of Social Media Users based on Behaviour and Language. Proceedings of the 38th European Conference on Information Retrieval (ECIR '16), pp. 689-695. doi:10.1007/978-3-319-30671-1_54

Please contact Prof. Xiaoming Fu [fu@cs.uni-goettingen.de](B/M/P)


New video/image encoding for DNN applications

  • New! Video/image encoding is important for image/video storage/delivery on Internet. It reduces file size by eliminating spatial-temporal redundancy. Along with the development of Deep Neural Network in the computer vision(CV) community, video/image encoding for DNN applications is becoming more and more crucial. This project attempts to compare the difference between video/image encoding for QoE and DNN applications; and explore the design space in the video/image encoding for DNN applications. We expect you have Digital Image Process and Computer Vision background, as well as programming skills like Python and C/C++.

Please contact Weijun Wang [weijun.wang@informatik.uni-goettingen.de] (B/M/P)

Super resolution technique for efficient video delivery

  • New! Super-resolution (SR) is one of the fundamental tasks in Computer vision. Video delivery on Internet or in WAN is important for various applications, eg., video analytics and video viewing. This project attempts to explore the potential of SR for video delivery. We expect you have Data Science and Computer Vision background, as well as programming skills like Python.

Please contact Weijun Wang [weijun.wang@informatik.uni-goettingen.de] (B/M/P)


Assessing city livability with big data

  • New! City livability is related to a number of factors, such as quality of life, job satisfaction, environment (green space, CO2/PM2.5, schooling/health support etc), policy, commuting time, entertainment. We utilize different data sources to understand their relation to the city livability, and analyze the coherent features which offer an evaluation framework for a city's attractiveness and livability for different types of citizens. We expect you have some statistics and machine learning background, as well as programming skills like Python.

Please contact Prof. Xiaoming Fu (B/M/P).

Socioeconomic analysis on commuters

  • New! Understanding the commuter behaviour and the factors that lead to commuting are more important today than ever before. With steadily increasing commuter numbers, the commuter traffic can be a major bottleneck for many cities. The increasing awareness of a good work-life balance leads to more people wanting shorter commuting distances. The commuter behaviour consequently plays an increasingly important role in city and transport planning and policy making. This topic aims to infer knowledge from commuter data, analyzing the influence of GDP, housing prices, family situation, income and job market on the decision to commute. We expect you have some statistics and machine learning background, as well as programming skills like Python.

Please contact Prof. Xiaoming Fu (B/M/P)


Socioeconomic Status and Internet Language Usage

  • New! Numerous people write social media posts and exchange messages with colleagues, friends, acquaintances or even strangers on different platforms. We would like to understand how the underlying social class membership (socioeconomic status) affects Internet users' language use, by investigating the sociolinguistic features in users' posts/messages across a multitude of datasets and their relationship to their socioeconomic status. We expect you have some statistics and textual analysis/natural language processing background, as well as programming skills like Python.

Please contact Prof. Xiaoming Fu (B/M/P)


Multimedia Resource Allocation for QoE Improvement by Deep Learning

  • New! Deep learning has been widely used in various real-time applications and systems. Dynamic resource allocation for multimedia (e.g. Video) to improve QoE is an interesting topic. We need three students for this topic. We expect you have a background in deep learning and computer network, as well as programming skills like Python and Go.

(1) one to realize and improve the system for video transmission and network configuration according to resource allocation policy;

  • You will use QUIC [1] protocol (Go language) to implement network allocation and place the server part on AWS/other clouds.

(2) one to implement the deep learning algorithm to design the controller for dynamic resource allocations.

(3) one student for the QoE model using deep learning.

Please contact Dr.Tingting Yuan [tingting.yuan@cs.uni-goettingen.de ] and Weijun Wang [weijun.wang@informatik.uni-goettingen.de](B/M/P)


-->

Ongoing Topics

Completed Topics

Topic Topic advisor Initial readings Description Student
Bio-Data analysis (Student project) Mayutan Arumaithurai Assigned to Lindrit
Sentiment Analysis (Student project) Hong Huang Assigned to Beatrice Kateule
Analysis of Business Transitions: A Case Study of Yelp (Bachelor Thesis) Hong Huang Assigned to Marcus Thomas Khalil
Understanding Group Patterns in Q&A Services (Bachelor Thesis) Hong Huang Assigned to Jonas Koopmann
COPSS-lite : Lightweight ICN Based Pub/Sub for IoT Environments (Master Thesis) Sripriya Assigned to Haitao Wang
A ICN Gateway for IoT (Bachelor Thesis) Sripriya Assigned to Janosch Ruff
Build a personalized context-aware recommender system for customers according to their own interest. Completed by Haile Misgna
Emotion Patterns Analysis in OSNs (Bachelor thesis Project) Hong Huang,Xu Chen We aim to study the emotion patterns in the Twitter service and predict the future emotion status of users. Completed by Stefan Peters
Implementation of a pub/sub system (Student project) Jiachen Chen Mayutan Arumaithurai The aim of the work is to show how application layer intelligence cupled with network layer pub/sub can be beneficial to both users as well as network operators Completed by Sripriya
Large Scale Distributed Natural Language Document Generation System (Student project at IBM) Mayutan Arumaithurai The work was done at IBM Completed by Eeran Maiti
Investigate real time streaming tools for large scale data processing (Student project) Mayutan Arumaithurai The aim of the work is to compare real time streaming tools. Completed by Ram
Software-Defined Networking and Network Operating System (Student project) Mayutan Arumaithurai SDN based ntwork operating system Completed by Rasha
GEMSTONE goes Mobile (BSc Thesis/Student Project) David Koll Portation of a Decentralized Online Social Network to the Android Platform Completed by Fabien Mathey and improved by Eeran Maiti
Transitioning of Social Graphs between Multiple Online Social Networks (BSc Thesis) David Koll Portation of friendship graphs between different Online Social Networks Completed by Kai-Stephan Jacobsen
Prevention and Mitigation of (D)DoS Attacks in Enterprise Environments (BSc Thesis) David Koll An analysis of enterprise infrastructures and their vulnerarbility towards attacks from the outside. Completed by David Kelterer
Sybils in Disguise: An Attacker View on OSN-based Sybil Defenses (Student Project and MSc Thesis) David Koll An analysis of fake detection approaches in social networks. Completed by Martin Schwarzmaier
Design and Implementation of a distributed OSN on Home Gateways (Student project and Master's Thesis) David Koll Completed by Dieter Lechler


QUIC or Multipath QUIC Design

  • New! Implement algorithms for improving QUIC or Multipath QUIC performance. (B/M/P, at least familiar with one programming language (eg. C++, go or Python).) Please contact Yali Yuan (Finished)


Segment Routing based SDN

  • NEW! Winter 2018/2019 There are many topics opened for Master and Bachelor theses and projects. Please contact Osamah Barakat


Software Defined Networks (SDN)


Data Analysis with Bio data

  • NEW! 2019 ' if you are interested in topics in this area please get in contact with Mayutan Arumaithurai

Data Crawling and analysis

Massive Data Mining and Recommender System

  • if you are interested in other topics in this area please get in contact with Hong Huang

Social Networking(finished)


  • For a full list of older topics please go here.
Retrieved from "https://wiki.net.informatik.uni-goettingen.de/index.php?title=Theses_and_Projects&oldid=7721"