Commit a920225527d02bf6cff21d1aab324acbf708af75

1 parent
0bfb7fb17d

Exists in
master

### Removed red color, except for 3 things

Showing
**3 changed files**
with
**37 additions**
and
**37 deletions**
Side-by-side Diff

CHIPLAYpaper/MarketPaper.log
View file @
a920225

1 | -This is pdfTeX, Version 3.1415926-2.5-1.40.14 (MiKTeX 2.9) (preloaded format=pdflatex 2015.4.9) 29 JUL 2016 02:20 | |

1 | +This is pdfTeX, Version 3.1415926-2.5-1.40.14 (MiKTeX 2.9) (preloaded format=pdflatex 2015.4.9) 29 JUL 2016 18:45 | |

2 | 2 | entering extended mode |

3 | 3 | **MarketPaper |

4 | 4 | (C:\Users\Olivier\gitlab\market-game\CHIPLAYpaper\MarketPaper.tex |

... | ... | @@ -2019,7 +2019,7 @@ |

2019 | 2019 | LaTeX Font Info: Font shape `OT1/txss/m/n' will be |

2020 | 2020 | (Font) scaled to size 7.59998pt on input line 42. |

2021 | 2021 | |

2022 | -Underfull \vbox (badness 10000) has occurred while \output is active [] | |

2022 | +Underfull \vbox (badness 3601) has occurred while \output is active [] | |

2023 | 2023 | |

2024 | 2024 | pdfTeX warning (ext4): destination with the same identifier (name{page.}) has b |

2025 | 2025 | een already used, duplicate ignored |

... | ... | @@ -2050,7 +2050,7 @@ |

2050 | 2050 | Here is how much of TeX's memory you used: |

2051 | 2051 | 32025 strings out of 493921 |

2052 | 2052 | 643088 string characters out of 3144871 |

2053 | - 709988 words of memory out of 3000000 | |

2053 | + 709971 words of memory out of 3000000 | |

2054 | 2054 | 34443 multiletter control sequences out of 15000+200000 |

2055 | 2055 | 71121 words of font info for 278 fonts, out of 3000000 for 9000 |

2056 | 2056 | 841 hyphenation exceptions out of 8191 |

... | ... | @@ -2093,7 +2093,7 @@ |

2093 | 2093 | Files (x86)/MiKTeX 2.9/fonts/type1/urw/times/utmbi8a.pfb><C:/Program Files (x86 |

2094 | 2094 | )/MiKTeX 2.9/fonts/type1/urw/times/utmr8a.pfb><C:/Program Files (x86)/MiKTeX 2. |

2095 | 2095 | 9/fonts/type1/urw/times/utmri8a.pfb> |

2096 | -Output written on MarketPaper.pdf (12 pages, 364815 bytes). | |

2096 | +Output written on MarketPaper.pdf (12 pages, 364768 bytes). | |

2097 | 2097 | PDF statistics: |

2098 | 2098 | 549 PDF objects out of 1000 (max. 8388607) |

2099 | 2099 | 106 named destinations out of 1000 (max. 500000) |

CHIPLAYpaper/MarketPaper.pdf
View file @
a920225

CHIPLAYpaper/MarketPaper.tex
View file @
a920225

... | ... | @@ -160,7 +160,7 @@ |

160 | 160 | %between the number of participants in a crowdsourcing group and the difficulty of the problem in order to find optimal size of groups [ref], and also the possible reactions |

161 | 161 | %between competing crowdsourcing groups that can attack and sabotage each other [ref]. |

162 | 162 | |

163 | -{\color{blue} Computer games are now a widely used and effective media to complete crowdsourcing tasks \cite{krause2015play}. Multiple studies already investigated the impact of design on human-computing tasks \cite{Krause:2010:FPE,ipeirotis2014quizz,kaufman2016investigating}. Incidentally, they are also an excellent and potentially very powerful framework to study collaborative processes.} | |

163 | +Computer games are now a widely used and effective media to complete crowdsourcing tasks \cite{krause2015play}. Multiple studies already investigated the impact of design on human-computing tasks \cite{Krause:2010:FPE,ipeirotis2014quizz,kaufman2016investigating}. Incidentally, they are also an excellent and potentially very powerful framework to study collaborative processes. | |

164 | 164 | |

165 | 165 | Historically, computation on graphs has proven to be a good model to study the performance of humans in solving complex combinatorial problems \cite{Kearns:2006aa}. Experiments have been conducted to evaluate the dynamics of crowds collaborating at solving graph problems \cite{DBLP:journals/cacm/Kearns12} but still, little is known about the efficiency of the various modes of interaction. |

166 | 166 | |

167 | 167 | |

... | ... | @@ -192,9 +192,9 @@ |

192 | 192 | |

193 | 193 | \section{Problem} |

194 | 194 | |

195 | -The game was implemented to solve a graph problem, which is the problem of finding maximal cliques in a multigraph. {\color{red} In simpler terms, | |

195 | +The game was implemented to solve a graph problem, which is the problem of finding maximal cliques in a multigraph. In simpler terms, | |

196 | 196 | the goal is to find subsets of vertices in the graph such that any two distinct vertices in the subsets are connected. In the case of this problem, we are dealing with |

197 | - a multigraph, which is simply a graph that can have multiple parallel edges.} | |

197 | + a multigraph, which is simply a graph that can have multiple parallel edges. | |

198 | 198 | Let $G(V,E)$ be a multi-colored graph, where each vertex $v \in V$ has a set of colors $c(v)$. There is a colored edge $e=(v,u) \in E$ |

199 | 199 | between the vertices $v$ and $u$ for every color in $c(v) \cap c(u)$ ({\em i.e.}, one for every color that they have in common). In other words, there is |

200 | 200 | no colored edge between two vertices $v$ and $u$ for which $c(v) \cap c(u) \neq \emptyset$. Let $|C|$ be the total number of colors in the graph. |

201 | 201 | |

202 | 202 | |

203 | 203 | |

... | ... | @@ -445,22 +445,22 @@ |

445 | 445 | \begin{enumerate} |

446 | 446 | \item Percentage of the problem solved |

447 | 447 | \item Total experience points earned by the players |

448 | -\item Average sequence length of the sequences created by the players {\color{red}($\sum_{i=1}^{TS} sL_i / TS$)} | |

449 | -\item Average number of colors in common of the sequences created by the players {\color{red}($\sum_{i=1}^{TS} nbC_i / TS$)} | |

450 | -\item Proportion of sequences of more than one color in common created by the players {\color{red}($(TS - T1C) / TS$)} | |

448 | +\item Average sequence length of the sequences created by the players ($\sum_{i=1}^{TS} sL_i / TS$) | |

449 | +\item Average number of colors in common of the sequences created by the players ($\sum_{i=1}^{TS} nbC_i / TS$) | |

450 | +\item Proportion of sequences of more than one color in common created by the players ($(TS - T1C) / TS$) | |

451 | 451 | \item Number of circles sold individually to another player |

452 | 452 | \item Number of sequences bought from other players (buyouts) |

453 | 453 | \end{enumerate} |

454 | - {\color{red} where $TS$ is the total number of sequences sold by all players during a session, $sL_i$ is the sequence length of sequence $i$, | |

455 | - $nbC_i$ is the number of colors in common of sequence $i$ and T1C is the total number of sequences of one color sold by all players during a session.} | |

454 | + where $TS$ is the total number of sequences sold by all players during a session, $sL_i$ is the sequence length of sequence $i$, | |

455 | + $nbC_i$ is the number of colors in common of sequence $i$ and T1C is the total number of sequences of one color sold by all players during a session. | |

456 | 456 | |

457 | 457 | \subsection{Game sessions} |

458 | 458 | |

459 | 459 | We recruited 120 people in total to test our game. Out of those 120 participants, 35\% were female and 65\% were male. Their average age was 25.7 and average video gaming time |

460 | 460 | per week was 4.7 hours. |

461 | -{\color{red} Roughly 75\% of the participants were undergrad students from McGill University and University of Montreal. | |

461 | +Roughly 75\% of the participants were undergrad students from McGill University and University of Montreal. | |

462 | 462 | The other 25\% were grad students and volunteers recruited through social media. Out of all the participants, about half of them were currently studying or working |

463 | - in computer science.} | |

463 | + in computer science. | |

464 | 464 | |

465 | 465 | We divided the participants into groups of 10 and repeated three times each of the four |

466 | 466 | game conditions presented in the previous subsection. |

... | ... | @@ -472,7 +472,7 @@ |

472 | 472 | %coming from the experience gained by the players if they played a second time. |

473 | 473 | Before starting each game session, the players were shown a document explaining |

474 | 474 | the rules of the game and the interface. |

475 | -{\color{red} The participants were told that they were playing a human computing game that aims to solve a graph problem, but the mathematical problem in question was not described to the players.} | |

475 | +The participants were told that they were playing a human computing game that aims to solve a graph problem, but the mathematical problem in question was not described to the players. | |

476 | 476 | %\sout{(to keep the instructions as simple as possible), only the rules of the game.}} |

477 | 477 | They were also asked to fill in a questionnaire so that we could get information on the participants, such as their age, |

478 | 478 | their abilities at puzzle solving and their experience with video games for example. For all the experiments, the game session lasted 45 minutes. |

... | ... | @@ -869,10 +869,10 @@ |

869 | 869 | \caption{Average statistics on the top 12 players vs the others.}\label{tab_playerStats} |

870 | 870 | \end{table} |

871 | 871 | |

872 | -{\color{red} We also looked at the mean age, mean video game time per week, and mean puzzle solving self evaluation for each of the 12 groups. We used | |

873 | - linear regression to measure the correlation between those mean values and the total experience points gained by all the players of the groups. Mean age and | |

874 | - mean video game time per week have a moderate level of correlation with total experience ($r = -0.67$ and $r = 0.58$ respectively), while mean | |

875 | - self evaluation only exhibits a weak level of correlation with total experience ($r = 0.39$).} | |

872 | +We also looked at the mean age, mean video game time per week, and mean puzzle solving self evaluation for each of the 12 groups. We used | |

873 | +linear regression to measure the correlation between those mean values and the total experience points gained by all the players of the groups. Mean age and | |

874 | +mean video game time per week have a moderate level of correlation with total experience ($r = -0.67$ and $r = 0.58$ respectively), while mean | |

875 | +self evaluation only exhibits a weak level of correlation with total experience ($r = 0.39$). | |

876 | 876 | |

877 | 877 | \section{Conclusion} |

878 | 878 | |

879 | 879 | |

... | ... | @@ -908,24 +908,24 @@ |

908 | 908 | have a strong self evaluation of their puzzle solving skills are able to understand the rules |

909 | 909 | of the game and find winning strategies faster than the average participant. |

910 | 910 | |

911 | -{\color{red} In the context of a human computing game, where participants are contributing for free, players' perception and statisfaction with the gameplay is very important. | |

912 | - We asked the players in the questionnaire to rate their experience with the game %on a scale of 1 to 10 | |

913 | - and the average score was 71.6\%. | |

914 | - We also gathered some feedback when we were talking to the participants after the sessions. | |

915 | - %As mentioned earlier, people who played the game with the market enjoyed it a lot more than the ones who played without it. | |

916 | - Since the game has no end in itself (we decided to limit the sessions to 45 minutes), some players mentioned that it is very addictive. | |

917 | - %\sout{(some wanted us to release the game publicly so they could play at home)}. | |

918 | - Competing against other players to be at the top of the leaderboard was also a powerful | |

919 | - motivation for the players. Obviously, we also had some negative comments: some players found the game hard to understand, while some very good players found it too easy. | |

920 | - Overall, it is quite interesting to see that players enjoyed the game in its current state without any extensive effort put into visual aesthetics. | |

911 | +In the context of a human computing game, where participants are contributing for free, players' perception and statisfaction with the gameplay is very important. | |

912 | +We asked the players in the questionnaire to rate their experience with the game %on a scale of 1 to 10 | |

913 | +and the average score was 71.6\%, which is significantly above average. | |

914 | +We also gathered some feedback when we were talking to the participants after the sessions. | |

915 | +%As mentioned earlier, people who played the game with the market enjoyed it a lot more than the ones who played without it. | |

916 | +Since the game has no end in itself (we decided to limit the sessions to 45 minutes), some players mentioned that it is very addictive. | |

917 | +%\sout{(some wanted us to release the game publicly so they could play at home)}. | |

918 | +Competing against other players to be at the top of the leaderboard was also a powerful | |

919 | +motivation for the players. Obviously, we also had some negative comments: some players found the game hard to understand, while some very good players found it too easy. | |

920 | +Overall, it is quite interesting to see that players enjoyed the game in its current state without any extensive effort put into visual aesthetics. | |

921 | 921 | |

922 | - Another important point for a collaborative game like ours is scalability. In the tests that we have done during the development of the game, | |

923 | - we have noticed that having more players is actually beneficial to the system. | |

924 | - There are more sellers and buyers on the market, which results in a market that is a lot more dynamic (bidders get their circles faster, prices change more often, etc.). | |

925 | - A more dynamic market allows different strategies to be equally profitable (e.g. some players can decide to focus on the Master Trader skill that boosts prices for | |

926 | - selling circles, and then focus entirely on selling individual circles to other players). | |

927 | - We are planning as future work to build an online version of the game that would be available 24/7. | |

928 | - This will allow us to analyze the differences in gameplay and quality of solutions between peak hours (with possibly thousands of players at the same time) and normal hours.} | |

922 | +Another important point for a collaborative game like ours is scalability. In the tests that we have done during the development of the game, | |

923 | +we have noticed that having more players is actually beneficial to the system. | |

924 | +There are more sellers and buyers on the market, which results in a market that is a lot more dynamic (bidders get their circles faster, prices change more often, etc.). | |

925 | +A more dynamic market allows different strategies to be equally profitable (e.g. some players can decide to focus on the Master Trader skill that boosts prices for | |

926 | +selling circles, and then focus entirely on selling individual circles to other players). | |

927 | +We are planning as future work to build an online version of the game that would be available 24/7. | |

928 | +This will allow us to analyze the differences in gameplay and quality of solutions between peak hours (with possibly thousands of players at the same time) and normal hours. | |

929 | 929 | |

930 | 930 | \section{Acknowledgments} |

931 | 931 |