CN104318068A - Competitive sports teamwork mode analogy method based on chordal graph visualization - Google Patents

Abstract

The invention discloses a competitive sports teamwork mode analogy method based on chordal graph visualization. The method comprises the following steps: 1) establishing a foundation database of game related information; 2) constructing a chordal graph according to the assisting information of players in a single game live broadcast record, i.e. dividing each player into a scoring sector and an assisting sector, respectively representing a scoring record and an assisting frequency record of each player, connecting the corresponding assisting sectors and scoring sectors of both cooperated parties by a chord so as to obtain a detail record of a game; 3) clustering the players with a multidimensional scaling method according to the historical record data of mutual assisting information of players in a team to obtain the two-dimensional coordinates of m vectors so as to obtain the clustering information of m players. Compared with the prior art, the competitive sports teamwork mode analogy method has the characteristics that chordal graph visualization design is adopted so as to bring convenience for users to better know and analyze game score distribution and player cooperation situations.

Description

Based on string figure visual athletics sports team matching model analogy method

Technical field

The present invention relates to the visual analysis technical field of information, the score mating information using string figure to contrast between ball in play person carries out visual presentation and interaction analysis, simultaneously by the cooperation tendency between Multidimensional Scaling methods analyst team sportsman, and then effectively study the tactics framework of team.

Background technology

In sports team sporting contests, due to game plan and cooperation tendency, often can be cooperatively interacted between part sportsman score, these matching models are very important for attacking and defending both sides, tactics adjustment can be made targetedly by the matching model research of one's own side, the associative mode studying the other side then can effectively see clearly tactics framework, and then arranges suitable Defending Policy, and therefore research team's Cooperation Strategy and pattern have broad application prospects.

Visual analysis is the emerging technology grown up in recent years, it is the product deriving from scientific visualization and information visualization field and combine closely with application background, can help user's cognitive complex data find data pattern behind more efficiently and accurately, user can carry out visual analysis to solve corresponding practical problems to data easily and quickly in conjunction with interactive operation simultaneously.

So far, although people around athletics sports match in matching model analysis and research expand extensive work.Based on match video, people can extract the Context event in video, conclude, and can coordinate Event Distillation displaying, but display form is more obscure, simultaneously can lack part information, and the cooperation details between sportsman but cannot represent effectively; By the pattern in match historical information is analyzed and is processed, people can carry out exploratory development to the technique and tactics of team, but exhibition method is not too directly perceived, and user clearly cannot know matching model; In addition, pass, by sportsman is considered as node, is considered as line by somebody, thus connects team for network, but score information but cannot show effectively.

Do not occur that one intuitively, effectively can show mating information yet at present, simultaneously according to the method for matching model Modeling Research team repertoire.

Summary of the invention

In order to overcome limitation and the deficiency of above-mentioned prior art, the present invention proposes a kind of based on string figure visual athletics sports team matching model analogy method, in conjunction with visual analysis and statistical study, mating information in match score is fully shown, and coordinate data to draw the matching model of team based on history, thus the repertoire of team is analysed in depth from different scale, for relevant Decision, person provides technical support.

The present invention proposes a kind of based on string figure visual athletics sports team matching model analogy method, the method comprises the following steps:

The basic database of step one, foundation match relevant information, comprise and contain secondary attack promoter and to compete live record with the single game of secondary attack terminator, and, in team, between sportsman, the mutual secondary attack historical data of every game is also obtained by official website, thus obtains coordinating sum between the sportsman within a racing season;

The secondary attack information architecture string figure of sportsman in step 2, live record of competing according to single game, that is: each sportsman is divided into score sector and secondary attack sector, represent its call and assists record respectively, use string to be connected coordinating secondary attack sector corresponding to both sides with score sector, thus draw the detailed record of match;

Step 3, according to the historical record data of mutual secondary attack information between sportsman in team, use Multidimensional Scaling method to carry out cluster to sportsman, under theorem in Euclid space, build m vector represent m sportsman, satisfied energy-optimised equation below

minΣ _i＜j(||X _i-X _j||-δ _i，j) ²：

Minimum convergence result, that is: the two-dimensional coordinate of m vector, thus draws the clustering information of m sportsman;

Wherein, the distinct matrix of sportsman's distance is expressed as

$\left(\begin{array}{cccc}{\mathrm{\δ}}_{\mathrm{1,1}}& {\mathrm{\δ}}_{\mathrm{1,2}}& \·\·\·& {\mathrm{\δ}}_{1,m}\\ {\mathrm{\δ}}_{\mathrm{2,1}}& {\mathrm{\δ}}_{\mathrm{2,2}}& \·\·\·& {\mathrm{\δ}}_{2,m}\\ \·\·& \·\·\·& \·\·\·& \·\·\·\\ {\mathrm{\δ}}_{m,1}& {\mathrm{\δ}}_{m,2}& \·\·\·& {\mathrm{\δ}}_{m,m}\end{array}\right)$

M is the sum of sportsman, and n is the sum of statistics match, represent the sum of secondary attack mutually in the match of kth field between sportsman i and sportsman j; δ in distinct matrix _{i, j}be expressed as

Wherein, δ _irepresent certain sportsman i in team, δ _{i, j}represent the distance between sportsman i and sportsman j.In described step 2, if must be divided into autonomous singles' score, then its secondary attack sector is connected with the score sector of self.

The width of described string is directly proportional with coordinating the summation producing mark between both sides, and the number of times that both sides coordinate is more, and total score is more, and the width of string is also larger.

Compared with prior art, beneficial effect of the present invention can be summarized as following some:

1, use string figure the visual design, facilitate user to understand better and analyze the distribution of match score and sportsman's mated condition;

2, filter irrelevant information by interaction analysis, improve user to the reading of key message and cognitive ability;

3, incorporate the statistical study of historical statistical information, improve and the accuracy researched and analysed and comprehensive is coordinated to team.

Accompanying drawing explanation

Fig. 1 is Bulls's score of the embodiment of the present invention and secondary attack string figure;

Fig. 2 is Heat's score of the embodiment of the present invention and secondary attack string figure;

Fig. 3 is the Bulls lead man Luo Si score of the embodiment of the present invention and secondary attack string figure;

Fig. 4 is the Heat lead man James score of the embodiment of the present invention and secondary attack string figure;

Fig. 5 is that the Heat sportsman in conjunction with historical information of the embodiment of the present invention coordinates Multidimensional Scaling figure;

Fig. 6 is a kind of overall flow schematic diagram based on string figure visual athletics sports team matching model analogy method of the present invention.

Embodiment

Below in conjunction with the drawings and specific embodiments, the present invention is described in detail, but practical range of the present invention is not limited thereto.

Be that embodiment is also further elaborated the present invention by reference to the accompanying drawings below with NBA.

Of the present invention based on Multidimensional Scaling chord figure visual athletics sports team cooperation research method, comprise the following steps:

The basic database of step one, foundation match relevant information, database information derives from the live record of NBA official website, the live record of secondary attack promoter and secondary attack terminator (i.e. goal scorer) is contained in record, simultaneously, in team, between sportsman, the mutual secondary attack historical data of every game is also obtained by official website, thus obtains coordinating sum between the sportsman within a racing season.

Step 2, according to the secondary attack information architecture string figure of sportsman in single game match, each sportsman represents with the sector of different colours, and each sportsman is divided into again score sector and secondary attack sector, represent score source and secondary attack whereabouts respectively, string is used to be connected coordinating secondary attack sector corresponding to both sides with score sector, if autonomous singles' score must be divided into, then its secondary attack sector is connected with the score sector of self, the width of string is directly proportional with coordinating the summation producing mark between both sides, the number of times that both sides coordinate is more, total score is more, the width of string is also larger, thus draw the details of match score.Simultaneously user can also filter the information such as the secondary attack whereabouts of checking some sportsmen and score source by mutual, more clearly the team of this sportsman cognitive coordinate with field on contribute.

Step 3, historical record data according to cooperatively interact between sportsman in team (secondary attack), use Multidimensional Scaling method to carry out cluster to sportsman, δ _irepresent certain sportsman i in team, δ _{i, j}represent the distance between sportsman i and sportsman j, the number of times of assisting between these distance and both sides is inversely proportional to, and secondary attack total degree is more, distance nearlyer (as shown in Equation 1),

M is the sum of sportsman, and n is the sum of statistics match, represent the sum of secondary attack mutually in the match of kth field between sportsman i and sportsman j, and then the distinct matrix (as shown in Equation 2) of sportsman's distance can be obtained.

$\left(\begin{array}{cccc}{\mathrm{\δ}}_{\mathrm{1,1}}& {\mathrm{\δ}}_{\mathrm{1,2}}& \·\·\·& {\mathrm{\δ}}_{1,m}\\ {\mathrm{\δ}}_{\mathrm{2,1}}& {\mathrm{\δ}}_{\mathrm{2,2}}& \·\·\·& {\mathrm{\δ}}_{2,m}\\ \·\·& \·\·\·& \·\·\·& \·\·\·\\ {\mathrm{\δ}}_{m,1}& {\mathrm{\δ}}_{m,2}& \·\·\·& {\mathrm{\δ}}_{m,m}\end{array}\right)---\left(2\right)$

Under theorem in Euclid space, build m vector represent m sportsman, by solving energy-optimised equation (as shown in Equation 3),

minΣ _i＜j(||X _i-X _j||-δ _i，j) ² (3)

Draw the two-dimensional coordinate of m vector, thus obtain the clustering information of m sportsman, the matching model of team also can be embodied.

Practicality of the present invention, science and accuracy will be confirmed further by embodiment below, choose the showing tremendous enthusiasm NBA racing season first match to bull on October 30th, 2013 in following instance and Heat sportsman matching model is research object.

Example 1, single game field player coordinate analysis

According to method of the present invention and step, construct the viewable design figure (as Suo Shi Fig. 1 to 5) that sportsman coordinates, each sportsman represents with the sector of different gray scale, and each sportsman comprises a secondary attack sector (" the sportsman name _ ast " of secondary attack score in figure represents) and a score sector (" the sportsman name _ score " of score in figure represents), the situation of its secondary attack of the former teammate score, the latter represents that it accepts the scoring event of teammate's secondary attack, as shown in Figure 1 and Figure 2.

Heat shown in Fig. 2 is compared with the Bulls shown in Fig. 1, resolving to of this visualization result: Heat

Interior lines sportsman's score is less.Heat's three interior lines sportsman Bo Shi (bosh), Haas Farnham (harslerm) and Anderson (anderson) PTS only have 31 points; And Bulls interior lines sportsman Bu Zeer (boozer) only people's score be just 31 points, another one main interior lines substitute Ji Busen (gibson) also obtained 9 points.From secondary attack part, the secondary attack performance of bull interior lines sportsman is also better than showing tremendous enthusiasm interior lines sportsman.In general, inside attack is the center of gravity of bull in this competition, and inside attack performance is also more outstanding.Although do not preponderate in interior lines, finally showing tremendous enthusiasm main cause of still winning the game is that its outside line sportsman is as James (james), Wei De (wade), the performance such as Qian Mosi (chalmers) is outstanding, Heat emphasizes the fluency that group coordinates simultaneously, individual autonomy singles score is relatively less, relief pitchers is as Allan (allen) in addition, Shane Battier (battier) and Cole (cole) also show excellence, especially James and Allan can score to be assisted again teammate's score, have activated the attack of team, and bull exactly lacks such sportsman, and the main performance relying on Bu Zeer and other starting lineups.

Rely on filter operation can analyze the performance of team lead man, Fig. 3 is Bulls lead man Luo Si (rose), Fig. 4 is Heat lead man James (james), can find out James be secondary attack or individual score is all better than Luo Si, therefore show outstanding as team leader James.

According to the inventive method, user can judge the score distribution that team competes and mated condition accurately and rapidly, significantly improves the efficiency of analytical work.

Example 2, to analyze based on sportsman's matching model of historical data

Choose the analysis that Heat player for the match carries out matching model, consider that single game competition data may exist accidental error, so choose Heat in the racing season and the code of other 29 teams as data supporting, if sportsman A and sportsman B coordinate more, then both sides are then relatively near or can be classified as same cluster at Multidimensional Scaling cluster middle distance.

By the correlation step of above-mentioned Multidimensional Scaling method, the dendrogram of Heat sportsman on two-dimensional space finally can be obtained, as shown in Figure 5.Can see in the drawings; James, Wei De, Bo Shi and Qian Mosi tetra-starting lineups often assist mutually cooperation; and Cole, Lewis (lewis), Beasley (beasley) and 4, Anderson relief pitchers to gather be a class; though Allan be relief pitchers its well connected first send out battle array and substitute battle array; Shane Battier and Haas Farnham are then relative isolated, among the overall co-ordination seldom participating in team.

Use the visual analysis method in the present invention, user can integrate historical information, analyzes the mated condition between sportsman more comprehensively, exactly, thus provides the foundation of analysis for the technique and tactics research of team.

Although above content is implemented to describe to invention has been with figure in conjunction with the embodiments; but the present invention is not limited only to above-described embodiment; foregoing is only illustrative and not restrictive; the person skilled of this area is under enlightenment of the present invention helps, and the distortion carried out when not departing from present inventive concept and essence also should belong within protection scope of the present invention.

Claims (3)

1., based on a string figure visual athletics sports team matching model analogy method, it is characterized in that, the method comprises the following steps:

The basic database of step one, foundation match relevant information, comprise and contain secondary attack promoter and to compete live record with the single game of secondary attack terminator, and, in team, between sportsman, the mutual secondary attack historical data of every game is also obtained by official website, thus obtains coordinating sum between the sportsman within a racing season;

The secondary attack information architecture string figure of sportsman in step 2, live record of competing according to single game, that is: each sportsman is divided into score sector and secondary attack sector, represent its call and assists record respectively, use string to be connected coordinating secondary attack sector corresponding to both sides with score sector, thus draw the detailed record of match;

Step 3, according to the historical record data of mutual secondary attack information between sportsman in team, use Multidimensional Scaling method to carry out cluster to sportsman, under theorem in Euclid space, build m vector represent m sportsman, satisfied energy-optimised equation below

minΣ _i＜j(||X _i-X _j||-δ _i，j) ²：

Minimum convergence result, that is: the two-dimensional coordinate of m vector, thus draws the clustering information of m sportsman;

Wherein, the distinct matrix of sportsman's distance is expressed as

$\left(\begin{array}{cccc}{\mathrm{\δ}}_{\mathrm{1,1}}& {\mathrm{\δ}}_{\mathrm{1,2}}& \·\·\·& {\mathrm{\δ}}_{1,m}\\ {\mathrm{\δ}}_{\mathrm{2,1}}& {\mathrm{\δ}}_{\mathrm{2,2}}& \·\·\·& {\mathrm{\δ}}_{2,m}\\ \·\·& \·\·\·& \·\·\·& \·\·\·\\ {\mathrm{\δ}}_{m,1}& {\mathrm{\δ}}_{m,2}& \·\·\·& {\mathrm{\δ}}_{m,m}\end{array}\right)$

M is the sum of sportsman, and n is the sum of statistics match, represent the sum of secondary attack mutually in the match of kth field between sportsman i and sportsman j; δ in distinct matrix _{i, j}be expressed as

Wherein, δ _irepresent certain sportsman i in team, δ _{i, j}represent the distance between sportsman i and sportsman j.

2. one as claimed in claim 1 is based on string figure visual athletics sports team matching model analogy method, it is characterized in that, in described step 2, if must be divided into autonomous singles' score, then its secondary attack sector is connected with the score sector of self.

3. as claimed in claim 1 or 2 a kind of based on string figure visual athletics sports team matching model analogy method, it is characterized in that, the width of described string is directly proportional with coordinating the summation producing mark between both sides, and the number of times that both sides coordinate is more, total score is more, and the width of string is also larger.