Seminar TK: Ubiquitous Computing

Transcription

1 Seminar TK: Ubiquitous Computing Seminar 4 CP, Summer Term 2014 Immanuel Schweizer schweizer@tk.informatik.tu-darmstadt.de Based on slides by Dr. Leonardo Martucci, Florian Volk

2 General Information What? Read and analyze current scientific publications Topics: Ubiquitous Computing Networking (Distributed Systems), Security, HCI Who? How? Select a topic and study it Write a short report Review other reports Present your report BSc, MSc and Diploma students from Computer Science Electrical Engineering and related areas Florian Volk, Telekooperation 2

3 General Information Why? Introduction to a research area Learn to read and analyze scientific material Present your evaluation Language? English Even though your advisor might speak German, your report has to be in English When? April 22 (today) Introduction Topic Presentation Tutorial: Working with Literature April 25 Topic Selection June 23 First version of your report (for the review) June 30 Deliverable of the reviews July 9 First version of your presentation July 14 Final version of your report July 16 (13:00; A126) Presentation of your work Meetings with your advisor (optional) Florian Volk, Telekooperation 3

4 5 Steps to Success No Read Literature enough Yes Write Report Peer review Correct Report Presentation 1. Pick a topic, read the provided literature and find more literature 2. Write an overview or state-of-the-art report 3. Peer-Review process Your report will be reviewed by a colleague (and by your advisor) You will review a colleague s report 4. Correct and improve your report following the reviewer s comments 5. Give a presentation on your report Florian Volk, Telekooperation 4

5 Evaluation and Grading You get 4 graded credit points for Your report: 4-5 pages IEEE transactions style paper (find templates on the course web page) Your participation in the review: Both active and passive Your presentation: 15 minutes + discussion Report 60 % Review 15 % Presentation 25 % You need to pass all parts! Florian Volk, Telekooperation 5

6 At a Glance 4 CP seminar with topics on Security, Privacy, and Trust Deadlines Topic Selection: 25th of April Report s 1st version: 23rd of June Review: 30th of June 1st Version of presentation 9th of July Report s final version: 14th of July Presentation: 16th of July /seminar-telekooperation-s3/ schweizer@tk.informatik.tu-darmstadt.de Florian Volk, Telekooperation 6

7 Seminar Topics

8 Architecture Description Languages Architecture Description Languages (ADL) are used by System Designer as a conceptual model Focus is on ADLs for distributed systems They allow: Rapid Prototyping Direct Execution or simulation Your task Survey current ADLs Categorize them based on their capabilities Florian Volk, Telekooperation 8

9 Privacy-preserving Publish/Subscribe Motivation Publish/Subscribe is a hot candidate NextGen Internet Distributed services Security & privacy is crucial requirement Confidential information Anonymous comm. Trust and authenticity Task Discover and survey new publications What it the SotA? What new approaches are out there? joerg.daubert@cased.de 9

10 Machine Learning in Anomaly Detection: Artificially intelligent techniques to detect intruders Overview: Machine Learning: Techniques for extracting knowledge from data. Anomaly Detection: Intrusion detection using machine learning tools :) Are you interested in learning more about this field of Artificial Intelligence? Goal: Explore different tools and algorithms already used by anomaly detection systems. Review what the latest papers are talking about. Find different and current algorithms for supervised, unsupervised, semi-supervised and reinforcement learning.

11 Remote Exploits: Techniques to detect and prevent them Overview: Exploit: Programs that take advantage of bugs or vulnerabilities in a system. Remote Exploits: Exploits that work remotely over a network. Multiple techniques exist to try to detect remote exploitation attempts. Intrusion Detection Systems (IDS) already use some of these technologies. Goal: Identify current techniques applied in modern Intrusion Detection Systems. Survey on state-of-the-art remote exploit detection technologies. carlos.garcia@cased.de

12 Topic: Speech-Recognition & Topic Recognition History: Two independent fields for enhancing ASR HCI Field vs. NLP-field (Diphones vs. Word-relations) Also other fields like robotics are active Task: Identify the active fields where ASR is enhanced by topic / domain identification Create an overview about the promising fields on paper level for supporting interdisciplinary activities Starting Paper: Sorbello Topic Recognition System, some further papers available Telecooperation 12

13 Data Correlation & Aggregation in Collaborative Intrusion Detection Systems IDS: Monitors a host or network and analyzes them for signs of intrusions manifested by malicious behavior or security policy violations Collaborative IDS (CIDS): A number of IDS monitors that collaborate by exchanging alert data Motivation: How to reduce the huge amount of alert data that is exchanged over the CIDS? Data correlation and aggregation techniques: similarity-based, attack scenario-based, etc. Goal: State-of-the-art of CIDSs data correlation & aggregation techniques manolis@cased.de

14 Defense Mechanisms of P2P Botnets Overview: Botnets: Infected machines/computers over the Internet (via Malwares) and turned into Bots/Zombies/Drones. P2P botnets are very resilient to take downs Recent botnets defend themselves upon recognizing they are being investigated/probed Goal: State-of-the-art survey on P2P Botnet s defense mechanisms and countermeasures shankar.karuppayah@cased.de 14

15 Overview on Topics 1. Architecture Description Languages (Jens) 2. Privacy-preserving Publish/Subscribe (Jörg) 3. Machine Learning in Anomaly Detection (Carlos) 4. Remote Exploits (Carlos) 5. Speech-Recognition & Topic Recognition (Stephan) 6. Data Correlation & Aggregation in Collaborative Intrusion Detection Systems (Manolis) 7. Defense Mechanisms of P2P Botnets (Shankar) Florian Volk, Telekooperation 15

16 How to work with Literature and write Scientific Material by Leonardo A. Martucci Sascha Hauke Florian Volk proudly presented and edited by Immanuel Schweizer

17 CONTENT What s a scientific publication? Finding (good) references Correct referencing Writing your own paper Reviewing papers *parts of this slide set are based on material provided by Guido Rößling

18 What s a scientific publication? Basically a message With scientific background Offer a new insight of a scientific problem (solution) OR a survey of a research field The message is a claim That needs to be evaluated AND validated Leonardo Martucci - Telecooperation 18

19 Books Survey (mostly) about a topic Journal Articles Quality mostly depends on the Journal Good Journal Good Article Sometimes articles are outdated Conferences and Symposia Quality is usually connected to the Conference Good Conference Good Paper The most recent research achievements Workshops Mostly for work in progress Good for discussing new ideas Standard and RFCs Types of Publications Leonardo Martucci - Telecooperation 19

20 References and Referencing Refer back to the original source of information For others to identify the foundations of your work Giving credit, when credit is due Not doing so is REALLY bad practice aka plagiarism Grundregeln der wissenschaftlichen Ethik am Fachbereich Informatik Leonardo Martucci - Telecooperation 20

21 What should I reference? Scientific publications Articles, papers, books Standards RFC, ITU, IEEE, W3C etc. + All other non-scientific sources Surveys Magazines Reports Can I reference Wikipedia? or any other online material? YES, but mind: not reliable (or stable) information sources Leonardo Martucci - Telecooperation 21

22 Writing a Scientific Publication 1. First, define the message Objective of your publication define the area of research 2. Read the related work Define the work around your work Finding out what has been done 3. Implement your idea Evaluate your idea Validate your idea 3. Survey the related work Evaluate differences Identify trade-offs Introduce your own classification 4. Write your publication Leonardo Martucci - Telecooperation 22

23 1. Your Work, Your Message Finding the message The most difficult part (!) Also, the creative one going beyond the state of the art! A message that needs science Scientific foundations + challenges can be found in the related work Leonardo Martucci - Telecooperation 23

24 2a. Related Work? Where? How? Related Work? Where? For the initial literature ask a researcher in the field it will give you a broad idea about the area Check publication repositories ACM Digital Lib IEEE Xplore Google Scholar Academic Search Conference directories Authors home pages Other sources from the reference lists REPEAT Leonardo Martucci - Telecooperation 24

25 2b. Related Work and Relevance Related Work Identify the relevant sources Evaluating the importance of a publication 1. Read the abstract Yes Good 2. Check the reference list Yes Good No 3. Read the conclusions 4. Read the rest Yes Good Next Paper Paper Read Related work will Compare your results against their results Be used as input for a survey Leonardo Martucci - Telecooperation 25

26 Referencing: doing it right A reference looks like this: authors title publisher date page number how was it published (proceedings) there are also other reference styles if you use LaTeX to write your report, have a look at BibTeX. Leonardo Martucci - Telecooperation 26

27 4. Write your Publication Always have a good paper structure Organize your ideas Organize your papers Define it BEFORE starting to add text Plan the content of each section Writing skills No one learns without doing it General Guidelines: Be concise (dt. prägnant ) Be precise (dt. präzise ) Leonardo Martucci - Telecooperation 27

28 Peer-reviews Peer-reviews Peers review your work and verify its general quality Evaluate the work before being published Offer suggestions to improve the work (!) How s quality defined in a publication?* Novelty Soundness Evaluation + Validation Completeness Readability * it sometimes depends on the venue What to write Positive and negative aspects of the work Constructive criticism (if possible) Offer suggestions to improve the paper e.g. additional literature Suggest an overall evaluation of the work It is NOT the reviewer s work to correct the publication! to point typos (unless if it is one or two) Leonardo Martucci - Telecooperation 28

29 Summary A scientific publication is a message, a validated claim Refer to the original source of information, avoid plagiarism 1. Read the abstract Yes Good 2. Check the reference list 3. Read the conclusions Yes Yes Good Good No Next Paper 4. Read the rest Paper Read The peer-review should help, not criticize Leonardo Martucci - Telecooperation 29