Name of The Student Student ID Session 2. Present Address
Name of The Student Student ID Session 2. Present Address
Name of The Student Student ID Session 2. Present Address
(Thesis Proposal)
Date: 08-10-2018
Page 1 of 9
7. Introduction
Predicting an individual athlete’s performance based upon his/ her past record can be critical in
the selection of team members in international competitions. This process is highly subjective
and usually requires much expertise and negotiative decision making. In this study, we explore
the analysis of related commentary and user comments in such decision situations. We take the
case of team selection in international contests in cricket.
Cricket is a popular game played by a few countries. There are two versions of the game – Test
Cricket which is played over five days and one-day cricket, which is obviously played over a
day. One-day cricket was introduced in the English domestic season of 1963 due to the growing
demand for a shorter and more dramatic form of cricket to stem the decline in attendance. One-
day, single-innings matches often took place before this, but the innovation was the limiting of
each side’s innings to an agreed number of lovers (nowadays usually 50). The inaugural 1975
World Cup was a great success. The abbreviations ODI (One-day International) or sometimes
LOI (Limited Overs International) are used for international matches of this type. Frequent nail-
biting finishes and the impossibility of either side opting to play for a draw have seen ODI
cricket gain many supporters
8. Related Work
We have found that only a few studies have been published on rating players, forming teams
and analyzing commentaries.
Page 2 of 9
Duckworth and Lewis [1] have developed innovative rain interruption rules that are
extensively used in one-day cricket matches. Their methods differ from previous approaches in
that they take into account the available run scoring resources (overs and wickets) the two
teams have left. Overall, the more unused run-scoring resources a team at their disposal at the
end of an interrupted innings the more runs they would score if not interrupted.
S. R. Iyer & R. Sharda[2] did excellent work on rating players and player selection. They’ve
explored the use of neural networks to rate players and select specific players for a
competition.
9. Objectives
10. Methodology
In this thesis experiment, we’ll scrape commentary data from websites and analyze them to
score players on some simple rules.
The system architecture of the framework comprises 5 basic modules; Website Access Module,
Data Scraping Module, Data Storing Module, Player Ranking Module and Recommending
Module. The system architecture is showed in Figure 1.
In the Website access module first, we have to establish connection to get access from Website.
That connection will allow us to get access Website API. Then from Website API we need to
consider Rest client library. The Website API will give the pathway to get access tokens for
collecting data from Website.
Page 3 of 9
Website are deeply concern about privacy of their users. So as users can modify their privacy
settings. By using the access token which one we have gotten from access module, in the data
crawler module we can crawl user’s data through a java crawler.
In Data storing module, we will store our data separately after removing duplicate data if there
exists any. It will make our data more reliable. Along with crawl data, this module also handles
storage of important information for retrieval purpose.
In Data categorization module, we will sub-categorize videos based on tags. This process is
beneficiary for the recommendation module.
Page 4 of 9
Website Access Data Filtering
Developer
API Textual
Data
Match tag
HTML Scraping
Data Storing
Match with
categorized database
Remove
Duplicate
Data Database
Recommend
In the recommending module, this will be done through matching tag which we will extract
from his profile.
Page 5 of 9
The database initialization and processing module consists of some sub-modules: Parsing
through HTML texts, retrieving access token, data crawling and categorization, removing
duplicate data, storing data into database, tag extraction, recommending items to user. The
relationships among this sub modules are showed elaborately in Figure 2.
Website
Database
Scraper
No
No
Have Filtered?
access
token?
Yes
Yes Rank
Scrape Data
Line up
Page 6 of 9
11. Required Resources
The costs that will occur to implement our proposed system are estimated below:
13. References
Page 7 of 9
[1] F. C. Duckworth and A. J. Lewis, “A fair method for resetting the target in interrupted one-
day cricket matches”, J. Oper. Res. Soc. , 49, (1998), 220-227.
[2] S. R. Iyer, and R. Sharda, “Prediction of athletes performance using neural networks: An
application in cricket team selection”, Expert Systems with Applications, 36, 5510 (2009)
Page 8 of 9
Signature of the Supervisor
Page 9 of 9