Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Assessment Brief

Module title

User adaptive intelligent systems

Module code

COMP3771

Assignment title

Recommender System Prototype

Assignment type and description

Coursework

Rationale

The coursework will allow you to apply topics from the module in a practical context. The coursework will allow you to practice key employability skills: problem solving, creativity, product design, technical report writing.

Word limit and guidance

maximum of 1500 words and no more than 6 pages (excluding references)

Weighting

40% of the module grade

Submission deadline

23 November 2023

Submission method

Submit a pdf file in Gradescope following the link in MINERVA

Feedback provision

Feedback will be provided via Gradescope no later than 3 weeks after the submission deadline. Group feedback will be provided in a lecture, together with the module wrapup at the end.

Learning outcomes assessed

By completing the coursework, you will be able to:

· derive requirements for a user-adaptive system in a specific context;

· design and prototype a user-adaptive system using recommender techniques;

· broaden your knowledge of recommender systems by researching relevant literature;

· assess your proposed design for transparency and usability;

· assess the strengths and weaknesses of your proposed design;

· practice technical report writing.

Module leader

Prof. Vania Dimitrova

Other Staff contact

Dr. Mohammed Azhar Iqbal

1. Assignment guidance

Design a prototype of a personal wellbeing assistant to act as a “buddy” which helps people to have a healthy lifestyle and feel happy by offering personalised recommendations.

Your application should focus on a specific group of users. The report should clearly state the characteristics of the target user group to be supported by your proposed prototype. For example, you may consider students (you may need to specify the year or subject area), or people with special needs (e.g., people with disabilities, neurodivergent people, patients, people with caring responsibilities), or people from specific groups (e.g., children, elderly, people moving to a foreign country). Your report should provide a clear justification of the selected characteristics and why these characteristics bring personalisation needs for the personal wellbeing assistant.

2. Assessment tasks

Write a report presenting your design. The report should be maximum 1500 words and no more than 6 pages (excluding references).

Your report should be written in an academic style, and should include the following sections:

1. Introduction which includes:

· Justification of the selected target user group (justify with appropriate references why this group of people is selected for your application and why personalisation is needed).

· Outline of the requirements for your application by listing the requirements and describing how you came up with them (this can be based on references you have found to justify user needs or feedback, e.g. interviews/survey with  possible users).

2. Description of the recommender method that will be used, including:

· Justification of the selection of the recommender method using appropriate references to user-adaptive systems that use this recommender method (note that it is not necessary that the user-adaptive systems which have used the selected recommender method refer to wellbeing support.

· Description of the background data which will be used by the recommender method with appropriate illustrations (include description of the data and how the data will be collected or provided).

· Description of the input data which will be used by the recommender method with appropriate illustrations (include description of the data and how the data will be collected).

· Description of the recommender algorithm including a brief outline of the main steps in the algorithm.

3. Critical review of the proposed design, including:

· Strengths - describe two strengths of the proposed design and provide appropriate justification and illustration for each strength.

· Transparency - apply the checklist presented in the paper “Best Practices for Transparency in Machine Generated Personalization” by Schelenz, Segal, and Gal [paper link available in Minerva] – apply all five areas from Table 2 in the paper.

· Usability – familiarise with the usability challenges presented in the paper “Adaptive interfaces and agents” by Jameson. [pages 15-18, paper link available in Minerva] and identify two possible usability threats that are related to your proposed design, for each of these threats describe what appropriate preventive and remedial measures should be included.  

4. Video (3 minutes long) demonstrating your prototype, including:

· User scenario - the demo should be based on a user scenario that illustrates the user interaction with the systems.

· Data collection – the demo should show what data is collected from the user and how.

· User interface – the demo should show what information is presented to the user and should illustrate how the user will receive the recommendations.

· Online link to the video with a description how to access it.

Note that the demo should mock the user interface. You are free to use any software to develop your prototype. Low fidelity, e.g. storyboarding with PowerPoint or wireframes (e.g. balsamiq) will suffice for the task. If you prefer to use any high-fidelity prototyping, this will be fine too.

3. General guidance and study support

Resources

· Re-read articles issued in the module as you may find helpful ideas from these experts on the overall architecture, the design of the user model, and the user modelling methods to be used.

· Visit User Modeling Inc (https://www.um.org/) and Recommender Systems conferences (https://recsys.acm.org/) for examples of user-adaptive systems presented at past conferences. Reference the systems that inspire the personalisation features to include in your application. Your inspiration may come from another domain and can be adapted for the problem in this coursework.

· You should include references that come from scholarly outputs in adaptation and personalisation; for instance, the RecSys (Recommender Systems) or UMAP (User Modelling, Adaptation and Personalisation) conference series, the UMUAI (User Modeling and User-adapted Interaction) journal.  

Support

· The coursework brief will be presented in a lecture

· Use the module Teams space to ask questions about the coursework (and the module in general)

· Prof. Dimitrova’s office hour is Friday 16pm-17pm, room 2.01a, Bragg building.

· Dr. Iqbal’s will also have an office hour for the period 20 Oct-23 Nov, he will provide help specifically for the coursework (day/time of Dr. Iqbal’s office hour will be announced on Teams).

4. Assessment criteria and marking process

The report will be marked using the criteria below. The video of the prototype will be accessed via the link and instructions provided in the report.

Outline of the marking scheme:

Introduction                                                                                                               12 marks

Description of the recommender method 18 marks

Critical review 13 marks

Video presenting the prototype          21 marks

Write up 6 marks

Total 70 marks

5. Presentation and referencing

This coursework will be submitted as a written report.

The quality of written English will be assessed in this work. As a minimum, you must ensure:

· Paragraphs are used

· There are links between and within paragraphs although these may be ineffective at times

· There are (at least) attempts at referencing

· Word choice and grammar do not seriously undermine the meaning and comprehensibility of the argument

· Word choice and grammar are generally appropriate to an academic text

These are pass/ fail criteria. So irrespective of marks awarded else-where, if you do not meet these criteria you will fail overall.

6. Submission requirements

The report will be submitted as a pdf file via Gradescope. Follow the assessment submission link from the module Minerva space.

7. Academic misconduct and plagiarism

Leeds students are part of an academic community that shares ideas and develops new ones.

You need to learn how to work with others, how to interpret and present other people's ideas, and how to produce your own independent academic work. It is essential that you can distinguish between other people's work and your own, and correctly acknowledge other people's work.

All students new to the University are expected to complete an online Academic Integrity tutorial and test, and all Leeds students should ensure that they are aware of the principles of Academic integrity. 

When you submit work for assessment it is expected that it will meet the University’s academic integrity standards. 

If you do not understand what these standards are, or how they apply to your work, then please ask the module teaching staff for further guidance.

By submitting this assignment you are confirming that the work is a true expression of your own work and ideas and that you have given credit to others where their work has contributed to yours.

8. Assessment/ marking criteria grid

Detailed marking scheme

Section

Criteria

Marks available

Introduction

Selected target user group properly justified (3 marks);

Justification uses appropriate references (3 marks)

Appropriate method to derive requirements is used (3 marks)

Description of requirements (3 marks)

12

Description of the recommender method

The selection of the recommender method properly justified (3 marks)

The justification uses appropriate references to user-adaptive systems that use this recommender method (3 marks)

Background data properly described (4 marks)

Input data properly described (4 marks)

Appropriate description how background and input data will be used to produce recommendations (4 marks)

18

Critical review

Strength 1 properly described and illustrated (2 marks)

Strength 2 properly described and illustrated (2 marks)

Transparency check list properly applied (5 marks)

Usability threat 1 properly described and appropriate preventive measure suggested (2 marks)

Usability threat 1 properly described and appropriate preventive measure suggested (2 marks)

13

Video presenting the prototype

User scenario appropriate (3 marks)

The prototype demo shows clearly what data is collected about the user (3 marks)

The prototype demo shows clearly what information is shown to the user (3 marks)

The prototype demo shows clearly how the system adapts to the user (3 marks)

The prototype meets the requirements specified in the introduction (6 marks)

The demo properly links to the user scenario (3 marks)

21

Write up

Appropriate report structure (1 mark)

Appropriate formatting (1 mark)

Clear and coherent text (1 mark)

Grammatically correct language (1 mark)

Appropriate use of illustrations (1 mark)

Appropriate referencing (1 mark)

6

TOTAL

 

70