340 likes | 492 Views
Spring 2014. Getting Started with Research & The Literature Review Part – 2 Cont. Lecture Prepared by: Dr. Peter Bloodsworth Dr. Anjum Naveed & Dr. Usman Ilyas. This Week. The literature review – part 2 concluded Important tips Group activity Technical Writing – how to do it.
E N D
Spring 2014 Getting Started with Research&The Literature Review Part – 2 Cont Lecture Prepared by: Dr. Peter Bloodsworth Dr. AnjumNaveed & Dr. UsmanIlyas
This Week • The literature review – part 2 concluded • Important tips • Group activity • Technical Writing – how to do it
Write it Up • While doing Literature review, writing things down is important so that you: • Remember what you read • Understand what is written • Write in your own words • Can test your thinking • Any ideas that cross your mind while reading must be written. • Once written, ideas become far more clear than being in mind • Helps test the real worth of idea. • Try organizing whatever you have written, to give it form of a report. • Organizing the writing is a big problem. • We will talk of some of the difficulties and solutions at later stages
Write it Up • Be systematic • Put dates on pages of notes so when you look back later you know where you were in your thesis • Learn about referencing styles early on • Keep properly formatted references • This WILL save a lot of time later • Keep your notes safe – don’t lose them • Take them along when you meet your supervisor • Add details of such meeting into your notes • Don’t forget action points and deadlines – Regularly review how you are doing against them!!
Reading Group • Read in groups even if every member reads his/her own related material • Share what you found. • Try explaining what you understood to others • At times, knowledge of other domain can be highly related to yours. • If you can’t explain it then you need to revisit what you read. • Talking about your research and writing it down helps you: • Better understand it • Clear your mind • Identify areas where you are unclear yourself • Others may see things differently which can help
Reading Group • Some faculty members run paper discussions within their research groups. • Remember to share experiences with each other • Research is a collaborative process • Much of what we learn we get from others so share tips with each other • Search the web – so much excellent content out there • Remember we are all different and think in our own ways • Find methods which work for you and help you to understand what you are reading
Reading Group • When the literature review is complete you can break up your group right? • WRONG! • The literature review is a continuous process • You will need people to bounce ideas off • Support each other – research can be stressful (just ask anyone with a PhD ) • Having a fairly diverse group is ideal • Gives different perspectives to problems • Sometimes a fresh look from someone new to a topic can be a big help
Reading Group Activity Activity: Organise yourselves into provisional reading groups of 4-6 students – try to include people you don’t normally work with Each person in the group should: • Introduce themselves (name, e-mail address etc) • Say what topics currently interest them Identify a group coordinator to arrange regular meetings You have around 15 mins
Thesis Topics with Me 2014 • Indoor positioning using sensors and agent-based techniques. • Robotics (multi-rotor or quadcopter) control and decision making • Real-time control via agents in the cloud • Applications eg: disaster management, security.. • Heterogeneous robot teamwork using agent based methods • Intelligent download using the cloud • Social cloud computing – Facebook to share vms
What makes a Research Report • Claim – Problem Statement/Hypothesis/Idea • Argument for/against hypothesis • Possible solution to the problem • Supporting your claim • Analysis • Empirical (Has to be comprehensive and complete) • Theoretical
What makes a Research Report • Warrant the relevance of argument • Use of rules to link the reasoning with claims • Acknowledging and responding to anticipated objections and questions • Questioning your argument as the readers will • Considering alternatives to your argument • Strengthening your argument • Looking at problem/approach from all angles
What makes a Research Report • Evidence for soundness of your argument • Finally • Application, relevance to domain, significance and some history
Thesis Writing • Before you start: • Two methods : write as you go or write a the end • Set a target date for submission • Set intermediary goals and check points • Read up on the rules regarding the thesis • What format does it need to be in? • Is there a word limit? • Have you got everything you need to do the writing? • Do you still need to be doing testing / result gathering? • How will you balance this? • Does you supervisor have any time restrictions? Such as going to a conference or on leave for a period
A Good Thesis Looks Like… • Examples • Notice: • Good presentation • Well organised / structured • Good size not too short or too long • Follows the SEECS formatting requirements • Has been prepared carefully and has had time spent on it
An Example Structure • Abstract • Introduction • Previous work • Your method introduced • Your method perhaps in more detail • Implementation • Results • Conclusion and future work • Bibliography • Appendicies
Caution • Every supervisor may have their own preferred structure / way of thesis preparation • These are general guidelines which should apply in most cases • Communication with your supervisor is essential – try to have an open discussion • Don’t get upset if you don’t always like what you hear • This is for your benefit so accept advice • Query why you are asked to do something but never have heated discussions – will not help either of you
Abstract Two ways of approaching this: • Some like to do it first so that they have the structure / argument of the thesis settled • Others (me included) prefer to write this at the end once the main body of the thesis is finalised • I suggest doing both – create a draft at the beginning and revise it at the end • Try to make it concise yet interesting – want to grab the attention of the reader • Need to show why you have chosen a particular problem – “because my supervisor told me to” isn’t a good answer
Abstract Summarisevery briefly: • Why it is a worthy problem / potential impact • Why it hasn’t been solved so far • What you propose to do • How you are going to evaluate it • Summary of results • Try to keep within one page – approx 500 words • Write mainly in the future tense: “This thesis will show….” • Use the third person … No I, we etc.
Introduction • Start by introducing the problem domain in more detail • If you have written a draft abstract use it to guide you – expand each sentence from the beginning part of the abstract into one or more paragraphs • Significance / application of your solution • Plan Sections / Sub-sections before writing • Describe any assumptions / simplifications that you have made • Have a clear problem statement or research hypothesis or both • Have some research questions which encompass the problem statement or allow you to answer the hypothesis
Introduction • Aim for a logical flow between research questions • Check that they are not: • Too ambitious • Too vague • Badly worded so as to be unclear or misleading • Irrelevant to your work • Questions which have a yes / no answer
Introduction • Next you need to describe at a high level how you are going to do the research • This is often called your research method • Try to be clear in what you write • Have a well structured approach to doing your work • Be logical and rigorous • Briefly describe how you plan to evaluate your work • Could be quantitatively or qualitatively or both
Introduction • Finally list any constraints / boundaries that you are setting for your work clearly • Explain why throughout – for each decision you make • You may want to introduce a criteria by which you can judge the success of your system / concept • Last thing: describe the structure of the remaining thesis
Previous Work • Otherwise known as your literature review, have said a lot about that already in previous weeksso won’t repeat Some structural points that may help: • Start off with overview – what broad areas do you need to consider and how do they relate? • In each main area which systems / approaches / results are of interest and pick out specific ones that are particularly relevant to your work and why? • What did they do, what didn’t they do and what conclusions can you draw for your work? Explain conclusions as you go. • Have a final discussion or concluding section which focuses on your area and leads into the following chapter
Lead In and Lead Outs • This is very important in making a thesis readable • The end of each chapter should lead into the next one. • The start of the new chapter should pick up the theme of the previous lead out • Each section should lead into the next • Each paragraph should also lead into the next in a logical way
Chapters About Your Work • Plan the chapter carefully, list the content that you want to present down on paper before writing • Organise your points in a logical flowing way • Start off with a GENTLE introduction to your approach • Remember that you will be an expert in the area of your thesis but the reader might need time to get your message or meaning • Use simple examples to show how complex processes work • Pick a case study and use it throughout your thesis to explain points • Describe technical details very carefully and don’t assume that the reader knows it already
Chapters About Your Work • Explain architectural points clearly • Give clear reasons for the important decisions that you have made in designing your work • If there are potential drawbacks to the approach then acknowledge them and suggest how they may be overcome • Gradually go into more detail don’t just jump in • Use technical language appropriately and in the right context • Link the technical details of your proposed system / architecture to the problems with existing work that you identified in your Previous Work chapter
Implementation • Be very careful when you describe your implemented system • Ask a friend to read your description to see if they think that using it they could build the system • Be sure that you explain implementation choices • Need to convince the reader that you have been rigorous in the way you have set up your prototype • Ensure that implementation decisions don’t influence your final results by biasing your experiments • Try to predict potential nasty questions and answer them in advance – will make your final defense easier
Results • Describe carefully what you are trying to show and how this relates to your hypothesis / problem statement • Explain how you are going to do this using a set of experiments / tests • Thoroughly describe each experimental set up • As you do think about what a reader could question and try to give an answer in advance • How have you ensured that all tests are fair? • What assumptions have you made and why?
Results • Look back to your literature review – how do other researchers in your domain test their systems / approaches? • What metrics are commonly used? • Are there any standard datasets that researchers use to evaluate their work? • Be careful with your technical language especially if mathematical or statistical where there is very little room for error • Look up a term if you don’t completely understand it – don’t guess
Results • Relate your final results to the research method that you described in your first chapter • Apply any criteria for success that you set at the beginning of your thesis • Use appropriate methods of illustrating results such as graphs and tables • Be careful with the presentation of any figures you use make sure they: • Have a clear title • Have a legend where appropriate • Are readable • Make the point you want them to • Are linked into the text and explained • Don’t have too many figures as they break the text up
Conclusion and Future Work Very important chapter don’t rush it! • Start by reviewing what you said you would do in your abstract and Introduction • Summarise the main points that you have covered in the thesis – give special focus to your experimental set up and results • Show areas where you have made a contribution and answer any research questions you have set yourself
Conclusion and Future Work • Make a judgementregarding your hypothesis / problem statement and explain how your results confirm this • Say what directions your work may take in the future • Your chance to say what you would have done if you weren’t constrained by time • You can speculate regarding how your work could be used by others in the future • Be sensible though don’t claim too much
Job Done Then? • Sadly not • Now go through and make sure it is consistent • Get the logical flow right • Remove loose ends that examiners could latch on to • Will likely require significant tweaks to each chapter • Finally do a last proof read format check and get ready to submit