Subject to the External Examiner’s approval and may be subject to change LUBS5990M

1 of 4

This question paper
consists of 4 printed
pages, each is identified
by the Code Number
LUBS5990M
© UNIVERSITY OF LEEDS
(Semester 2, 2021/2022)

Assessed Coursework

LUBS5990M
Machine learning in practice

100% Assignment



BACKGROUND

Financing is critical for high-technology startups to compete and succeed in the market.
Initial coin offers (ICOs) are a new way for startups to raise funding. As of January 2021,
5728 ICO projects had raised more than $27 billion. From an entrepreneur's perspective,
ICOs are extremely appealing because they provide funding at all stages with virtually no
transaction expenses. ICOs are particularly fascinating from an investor's perspective
because they may result in higher profits.

In an ICO, a startup solicits investment from online investors by issuing and
selling blockchain-secured 'coins' (a digital medium of value based on blockchain
technology). These coins, sometimes referred to as 'cryptocurrencies,' can be traded
between investors but may also have some utility when used in in the product or service
offered (e.g., in their digital Apps) by the startups.

The majority of ICOs follow an 'all-or-nothing' model, in which the startup establishes a
fundraising goal (for example, raise $1 million in 30 days). If they raise enough money to
accomplish their goal, the ICO will be regarded a success. Otherwise, the fundraising
effort will be in vain, and they will receive nothing. On their campaign page, each ICO
presents the project and team details. Here is an example of an ICO fundraising page:
https://icobench.com/ico/verasity

YOUR TASK

The aim of the assignment is for you to demonstrate:
• your ability to understand data
• your ability to prepare data for modelling and evaluation
• your ability to apply appropriate modelling techniques to the processed data
• your ability to validate and interpret performances of different models
• your ability to present your analyses in the form of a technical report

You have been provided with a dataset with attributes about the ICO projects from different
startups. For this coursework, you need to produce models predicting whether a
startup will reach its fundraising goal successfully through the ICO. The following
variables could be used as potential predictors:

Variable Definition
ID Unique ID number of the startup company
goal An indicator variable which is set as ‘Y’ if the startup achieved
their funding goal (raised funding successfully) otherwise ‘N’.
Subject to the External Examiner’s approval and may be subject to change LUBS5990M

2 of 4

startdate The starting date of the funding raising campaign
enddate The end date of the funding raising campaign
coinNum The number of coins to be issued. A startup company can freely
decide the number of coins to be issued. A higher coin number
offered usually means a lower price for each coin.
teamSize The number of team members for the fundraising project.
country_region The country or region of the company located in
categories The categories (sectors) of the fundraising project. One project
could belong to multiple categories.
overallrating The overall rating score for the quality of an ICO project, given by
investment experts, ranged from 1(very poor) to 5 (very good).
ratingTeam The rating score for the project team, given by investment
experts, ranged from 1(very poor) to 5 (very good).
ratingProduct The rating score for the product or service offered by the project,
given by investment experts, ranged from 1(very poor) to 5 (very
good).
platform The name of the blockchain platform for a fund raising project
based on.
acceptingCurrency
Num
The number of currencies that the fundraising company would
accept as an investment. A startup company may accept multiple
cryptocurrencies, most are cryptocurrencies such as Bitcoin or
Ether.
whitepaper An indicator variable set as 1 if the fundraising project provided a
white paper on their campaign page, otherwise 0.
video An indicator variable set as 1 if the fundraising project provided a
video on their campaign page, otherwise 0.
socialMedia The activity level of the company in social media during the
campaign period, ranged from 0(least active) to 3(most active)
Github An indicator variable set as 1 if the fundraising project provided
their official Github page (for sharing their source code) on their
campaign page, otherwise 0.
teamLinkedIn The percentage of team members who provided their LinkedIn
profile on the project campaign page, ranged from 0% (no one
provided that) to 100%(all members provided that).
teamPhotos The percentage of team members who provided their photos on
the project campaign page, ranged from 0% (no one provided
that) to 100%(all members provided that).
CEOPhoto An indicator variable set as 1 if the CEO of the company provided
a photo in the campaign page, otherwise 0.

You can create your own variables based on existing variables. For example, the duration
of the fundraising campaign can be obtained by using the start and end dates; whether a
fundraising company is located in the USA can be obtained from the country variable. You
may gain higher marks for using other possible and well-justified predictor variables. You
could even link the given dataset with external data such as overall economic data or
Bitcoin prices during the campaign period to create useful variables.

WHAT YOU SHOULD SUBMIT
The assignment submission should be a technical report with summary of the analyses
performed. Your report must have following sections:
Subject to the External Examiner’s approval and may be subject to change LUBS5990M

3 of 4

• Introduction (Business understanding)
o This is to set the context of this technical report by describing the business
context and the background of the task.
• Data Understanding
o Show evidence of understanding of the data e.g. distributions, visualisation
of the data, etc.
• Data Preparation
o Describe how you pre-process the raw data to get it ready for modelling in
next stage. E.g., how you handle missing values, incorrect values, and
outliers; how you transform the data format, etc.
• Modelling
o Describe the predictors and output variables for your model. Provide
evidence that you have investigated relationships among the predictors.
o Describe different classification techniques that you have considered and
selected for this task and explain why it was chosen.
• Evaluation
o Describe in detail how you have tested your models. To gain high marks,
you should show evidence of using different robust measures to evaluate
the modelling techniques you have selected.
• Conclusion (Deployment)
o This section includes findings and conclusions from your analysis.


SUBMISSION GUIDELINES
The analyses you submit should be within the word counts specified. You may however
submit as an appendix output in support of material in the analyses, which will not be
included in the word count. This allows those marking the assignment to check the output
against your interpretation.

The marks for the assignment are awarded on the basis of the main text only.

You can use any analytics tool to conduct the data processing and analytics work.

ADDITIONAL READINGS
Fisch, C., 2019. Initial coin offerings (ICOs) to finance new ventures. Journal of Business
Venturing, 34(1), pp.1-22.

Huang, W., Vismara, S. and Wei, X., 2021. Confidence and capital raising. Journal of
Corporate Finance, p.101900.

Those papers may provide you some background information about the ICO and the
provided dataset. You DO NOT have to follow the methodology or variables used in those
papers. The papers can be downloaded from Minerva.


Assignments should be a maximum of 3,500 words in length.

All coursework assignments that contribute to the assessment of a module are subject to a
word limit, as specified in the online module handbook in the relevant module area of the
MINERVA. The word limit is an extremely important aspect of good academic practice,
and must be adhered to. Unless stated specifically otherwise in the relevant module
Subject to the External Examiner’s approval and may be subject to change LUBS5990M

4 of 4

handbook, the word count includes EVERYTHING (i.e. all text in the main body of the
assignment including summaries, subtitles, contents pages, tables, supportive material
whether in footnotes or in-text references) except the main title, reference list and/or
bibliography and any appendices. It is not acceptable to present matters of substance, which
should be included in the main body of the text, in the appendices (“appendix abuse”). It is not
acceptable to attempt to hide words in graphs and diagrams; only text which is strictly
necessary should be included in graphs and diagrams.

You are required to adhere to the word limit specified and state an accurate word count on the
cover page of your assignment brief. Your declared word count must be accurate, and should
not mislead. Making a fraudulent statement concerning the work submitted for assessment
could be considered academic malpractice and investigated as such. If the amount of work
submitted is higher than that specified by the word limit or that declared on your word count,
this may be reflected in the mark awarded and noted through individual feedback given to you.

The deadline date for this assignment is Needs to be approved by the Assessment
Office. Confirmed date will be announced in Minerva and in the class.

An electronic copy of the assignment must be submitted to the Assignment Submission area
within the module resource on the Blackboard MINERVA website no later than 12:00:00 prompt
on the deadline date.

Faxed, emailed or hard copies of the assignment will not be accepted.

Failure to meet this initial deadline will result in a reduction of marks, details of which can be
found at the following place:
https://students.business.leeds.ac.uk/assessment/code-of-practice-on-assessment/coursework/

SUBMISSION
Please ensure that you leave sufficient time to complete the online submission process, as
upload times can vary. Accessing the submission link before the deadline does NOT constitute
completion of submission. You MUST click the ‘CONFIRM’ button before 12:00:00 for your
assignment to be classed as submitted on time, if not you will need to submit to the Late Area
and your assignment will be marked as late. It is your responsibility to ensure you upload the
correct file to the MINERVA, and that it has uploaded successfully.

It is important that any file submitted follows the conventions stated below:

FILE NAME
The name of the file that you upload must be your student ID only.
ASSIGNMENT TITLE
During the submission process the system will ask you to enter the title of your submission.
This should also be your student ID only.
FRONT COVER
The first page of your assignment should always be the Assessed Coursework Coversheet
(individual), which is available to download from the following location:
https://students.business.leeds.ac.uk/forms-guidance-and-coversheets/
STUDENT NAME
You should NOT include your name anywhere on your assignment
END