Page 39 - Fister jr., Iztok, and Andrej Brodnik (eds.). StuCoSReC. Proceedings of the 2016 3rd Student Computer Science Research Conference. Koper: University of Primorska Press, 2016
P. 39
ualization of cycling training
Dušan Fister Iztok Jr. Fister Iztok Fister
Univerza v Mariboru Univerza v Mariboru Univerza v Mariboru
Fakulteta za elektrotehniko,
Fakulteta za strojništvo Fakulteta za elektrotehniko, racˇunalništvo in informatiko
Smetanova 17, 2000 Maribor
Smetanova 17, 2000 Maribor racˇunalništvo in informatiko
dusan.fister@student.um.si Smetanova 17, 2000 Maribor iztok.fister@um.si
iztok.fister1@um.si
ABSTRACT consumption), analysing daily weight and measuring aver-
age sleep quality by measuring movement of the arm [5].
In the era of big data, a flood of information obtained from
pervasive computing devices is being available. Using lat- All those properties, which became accessible to every ath-
est data mining technologies, lots of information can also lete, lead to huge and complex online trackers, offering lots
be processed. Since many of them do not present any sig- of data. But these data should be correctly interpreted in
nificant meaning to users, information could be ranged in order to became useful information for an athlete. In line
so-called significant classes, where these are selected accord- with this, different applications for visualization in sports
ing to their significance. This paper presents an analysis of have been emerged. For instance, the more frequently used
data obtained during the cycling training sessions made by visualization tasks are described in [7], but these are espe-
wearable sports tracker and later visualization of their most cially suited for the team sports. The analysis of the cycling
significant parameters. Every sports training activity should data is well described in [3], while the advanced data-mining
be presented in a simple, self-speaking and understandable approach in [6]. The cycling training sessions and its plan-
figure, from which it is possible to deduce difficulty, strain, ning can be seen in [4].
effort, power, conditions and pace of the visualized sports
training session. This paper proposes visualization of cycling elements, where
athlete’s data obtained from an archive of the activity data-
Keywords sets are interpreted. Based on athlete’s effort, specific figures
are built, from which it is possible to obtain the most signif-
visualization, cycling, cycling elements, .TCX icant class information about the performed sports training
activity.
1. INTRODUCTION
The organization of the paper is as follows. In second chap-
A use of GPS sports trackers (trackers) for sports train- ter, basic elements of cycling training are described in de-
ing purposes in cycling is increased every day. More and tails. Third chapter offers background and description of
more athletes use trackers to measure, accompany and con- proposed visualization, while the fourth chapter deals with
trol data obtained during their trainings, as well as perform results. The paper ends with conclusions and outlines the
later analysis and online planning. Trackers are today em- directions for future work.
bedded into sports-watches (e.g. Garmin Connect, Strava,
Movescount, Endomondo, Sports tracker, etc.) or mobile 2. ELEMENTS OF CYCLING TRAINING
devices offering an upload of datasets with tracked data to AND VISUALIZATION BACKGROUND
the mentioned sports tracker producer websites. Every cy-
clist collects his activity datasets in a so called calendar, In this paper, we are focused on visualization of cycling
showing daily training sessions. training datasets. Cycling datasets are, as mentioned in
Chapter 1, created during a sports training session. Cycling
Sport-watches and mobile devices provide a brand-new ap- trackers usually record properties from the most significant
proach of training, not only for cycling, but also for running, class consisting of:
swimming, canoeing, hiking, roller skating, skiing, fitness,
and other sports. Some of the specialized trackers support • position,
additional functions, like summing the number of daily steps
and predicting calories burnt (therefore measuring the daily • distance,
• duration,
• velocity,
• altitude,
• temperature,
• heart rate and
StuCoSReC Proceedings of the 2016 3rd Student Computer Science Research Conference 39
Ljubljana, Slovenia, 12 October
Dušan Fister Iztok Jr. Fister Iztok Fister
Univerza v Mariboru Univerza v Mariboru Univerza v Mariboru
Fakulteta za elektrotehniko,
Fakulteta za strojništvo Fakulteta za elektrotehniko, racˇunalništvo in informatiko
Smetanova 17, 2000 Maribor
Smetanova 17, 2000 Maribor racˇunalništvo in informatiko
dusan.fister@student.um.si Smetanova 17, 2000 Maribor iztok.fister@um.si
iztok.fister1@um.si
ABSTRACT consumption), analysing daily weight and measuring aver-
age sleep quality by measuring movement of the arm [5].
In the era of big data, a flood of information obtained from
pervasive computing devices is being available. Using lat- All those properties, which became accessible to every ath-
est data mining technologies, lots of information can also lete, lead to huge and complex online trackers, offering lots
be processed. Since many of them do not present any sig- of data. But these data should be correctly interpreted in
nificant meaning to users, information could be ranged in order to became useful information for an athlete. In line
so-called significant classes, where these are selected accord- with this, different applications for visualization in sports
ing to their significance. This paper presents an analysis of have been emerged. For instance, the more frequently used
data obtained during the cycling training sessions made by visualization tasks are described in [7], but these are espe-
wearable sports tracker and later visualization of their most cially suited for the team sports. The analysis of the cycling
significant parameters. Every sports training activity should data is well described in [3], while the advanced data-mining
be presented in a simple, self-speaking and understandable approach in [6]. The cycling training sessions and its plan-
figure, from which it is possible to deduce difficulty, strain, ning can be seen in [4].
effort, power, conditions and pace of the visualized sports
training session. This paper proposes visualization of cycling elements, where
athlete’s data obtained from an archive of the activity data-
Keywords sets are interpreted. Based on athlete’s effort, specific figures
are built, from which it is possible to obtain the most signif-
visualization, cycling, cycling elements, .TCX icant class information about the performed sports training
activity.
1. INTRODUCTION
The organization of the paper is as follows. In second chap-
A use of GPS sports trackers (trackers) for sports train- ter, basic elements of cycling training are described in de-
ing purposes in cycling is increased every day. More and tails. Third chapter offers background and description of
more athletes use trackers to measure, accompany and con- proposed visualization, while the fourth chapter deals with
trol data obtained during their trainings, as well as perform results. The paper ends with conclusions and outlines the
later analysis and online planning. Trackers are today em- directions for future work.
bedded into sports-watches (e.g. Garmin Connect, Strava,
Movescount, Endomondo, Sports tracker, etc.) or mobile 2. ELEMENTS OF CYCLING TRAINING
devices offering an upload of datasets with tracked data to AND VISUALIZATION BACKGROUND
the mentioned sports tracker producer websites. Every cy-
clist collects his activity datasets in a so called calendar, In this paper, we are focused on visualization of cycling
showing daily training sessions. training datasets. Cycling datasets are, as mentioned in
Chapter 1, created during a sports training session. Cycling
Sport-watches and mobile devices provide a brand-new ap- trackers usually record properties from the most significant
proach of training, not only for cycling, but also for running, class consisting of:
swimming, canoeing, hiking, roller skating, skiing, fitness,
and other sports. Some of the specialized trackers support • position,
additional functions, like summing the number of daily steps
and predicting calories burnt (therefore measuring the daily • distance,
• duration,
• velocity,
• altitude,
• temperature,
• heart rate and
StuCoSReC Proceedings of the 2016 3rd Student Computer Science Research Conference 39
Ljubljana, Slovenia, 12 October