The domain-specific features are represented as a confrontation matrix of batting features and bowling features. T CTBOW L Bowler's Technical Confrontation Tensor ECMBAT Batsman's External Confrontation Matrix ECMBOW L Bowler's External Confrontation Matrix.
Introduction I n today’s world, as sports are getting more competitive than ever, players and teams are looking
Contributions of the Thesis
- Contribution 1: Representing Cricket Text Commentary Data
- Contribution 2: Mining Strength and Weakness Rules of Cricket Players Knowledge of players’ strengths and weaknesses is the key to team selection and strategy planningKnowledge of players’ strengths and weaknesses is the key to team selection and strategy planning
- Contribution 3: Mining Temporal Changes in Strength and Weakness Rules of Cricket PlayersRules of Cricket Players
- Contribution 4: Mining Strength and Weakness Rules of Cricket Players in the Presence of External Factorsin the Presence of External Factors
- Contribution 5: Visualization of Similar Players Based on their Strength and Weakness Rulesand Weakness Rules
We highlight some of the strength and weakness rules achieved for batsman Steve Smith and bowler Kagiso Rabada. RQ: What are the rules of strength and weakness for a player in the presence of external factors that affect the game.
Outline of the Thesis
Data Representation: The strength and weakness vectors of all batsmen and bowlers were obtained from their strength and weakness rules. Mining strength rules and cricketers' weakness rules in the presence of external factors.
Cricket
The batsman can play different strokes to hit the ball to different regions of the field (shot areas). If Team A leads by at least 200 runs after the second innings, the captain of Team A can order Team B (which enforces the follow-on innings) to bat in the next innings.
Cricket Data Sources
- Structured Data
In the first innings, the batting team sets the goal for the fielding team, and in the second innings, the fielding team (which is now the batting team) tries to reach the goal. A team's innings ends when (i) the team is all-out, (ii) the team's captain declares the innings, (iii) the team batting fourth scores the required number of runs to win, or (iv) ) the time before the match expires.
BACKGROUND
- Unstructured Data
- Comparison of Cricket Data
- Related Work
- Sports Data Mining
- Text Representation
- Short Text Representation
- Text Visualization
- Short Text Visualization
- Summary
Various data mining techniques have been successfully applied to sports data generated in recent years. 98] addressed the one-topic assumption constraint in the field of short text topic modeling.
Representing Text Commentary Data I n this chapter, we discuss the representation of text commentary data and its challenges. Refer
- Text Commentary Data
- Short Text Commentary
- External Factor Data
- Text Commentary Acquisition
- Text Commentary Processing
- Short Text Commentary Processing
- External Factor Data Processing
- Feature Extraction
- Technical Features
- External Features
- Confrontation Matrix Construction
- Technical Confrontation Matrix
- External Confrontation Matrix
- Summary
The rest of the text describes how the ball is delivered and how the batsman plays it. The sparsity is resolved by mapping the text comment unigram and bigram to these features only.
Mining Strength and Weakness Rules of Cricket Players
Computational Definition of Strength and Weakness
In definition 4.1, when the batting function corresponds to the attack and involves any of the bowling functions. In definition 4.1, when the batting function corresponds to batted and involves any of the bowling functions.
Learning Strength and Weakness Rules
- Batting Analysis through CA
- Bowling Analysis through CA
The process of obtaining a bowling function is to take the inner product of Fbatting function0 with each bowling vector of G0, i.e. the row vector of the selected batting function in hFm×2, Gn×2imatrix. The process of obtaining bowling function is to take the inner product of Fbatting function0 with each bowling vector of G0 ie. the row vector of the selected batting function in the hFm×2, Gn×2i matrix.
Visualization of Strength and Weakness Rules
In addition to the strength rule and the weakness rule, it is important to learn the other rules for the bowler that correspond to the opposing batsman's response, score, footwork, and fielding zone. The rest of the process remains the same with learning the rules of strength and weakness.
Experiments
- Batting Analysis - Steve Smith
- Bowling Analysis - Kagiso Rabada
Player move-in deliveries to square area of the field Table 4.2: Rules obtained from Steve Smith's Batting Analysis. Using the proposed method, batsman Steve Smith's biplots against fast bowlers and spin bowlers are obtained (see Figure 4.7).
Validation
- Extrinsic Validation
- Intrinsic Validation
The following is a transcription of selected parts of the video, together with the rules obtained by the proposed method. The lowest mean sum of squared residual is 0.15 for batsman Joe Root, which means that his training and test data sets are most similar.
Baseline Comparison
- Strength and Weakness Visualization using Wordclouds
- Strength and Weakness Association Rules
We provide computational definitions of the strengths and weaknesses rules and then use ARM to build player-specific rules based on the definitions. The results of the strengths and weaknesses analysis for batsman Steve Smith against all bowlers in Test matches are shown in Table.
Web Application
Discussion
We present the rules of strength and weakness (maximum three bowling traits in which batsmen have shown strength or weakness) of the top batsmen ranked by the ICC in Table 4.8. The quality of the opposing player can certainly be a factor affecting the rules of strength and weakness of the individual player.
Summary
To resolve this, we have set up batsmen's strength and weakness rules separately for fast bowlers and spin bowlers (Section 4.4.3). From this table, we can see the difference in strength and weakness rules for different players.
Mining Temporal Changes in Strength and Weakness Rules of Cricket Players
Technical Confrontation Tensor
Using these yearly T CMBAT, we construct a Technical Confrontation Tensor (T CTBAT) of size (19×12 ×number of years), where rows correspond to batting functions of the player, columns correspond to bowling functions of opposing players, and tubes correspond to to the time frame (per year) in which the batsman has played. Using these yearly T CMBOW L, we construct a Technical Confrontation Tensor (T CTBOW L) of size (19 × 12 × number of years), where rows correspond to batting characteristics of opponent batsmen, columns correspond to bowling characteristics of bowlers, and pipes correspond to the time frame (per year) in which the bowler has bowled.
Learning Temporal Changes in Strength and Weakness Rules
- Temporal Analysis of Batting through TWCA
- Temporal Analysis of Bowling through TWCA
These inner product values are plotted using line plots to visualize annual changes in the power rule against a particular bowling function. These inner product values are plotted using line plots to visualize annual changes in strength rules on a particular bowling function.
Experiments
- Year-wise Analysis of Batting - Steve Smith
- Year-wise Analysis of Bowling - Kagiso Rabada
Applying the proposed approach described in Section 5.2.2, the annual changes in the strength and weakness rule of bowler Kagiso Rabada on his different deliveries are obtained and presented as line plots (see Figure 5.5). Annual changes in Strength Rule (SR) and Weakness Rule (WR) for bowler Kagiso Rabada on his full-length deliveries are shown in Figure 5.5a.
Web Application
Given the additional bowling features that change over time, the proposed temporal analysis is able to learn the evolution of such features. A use case would be to learn the evolution of bowlers' strengths and weaknesses in the presence of external characteristics such as field conditions and weather conditions.
Discussion
Summary
A detailed investigation of quality as an additional parameter in temporal analysis of strength and weakness rules is left for future work.
Mining Strength and Weakness Rules of Cricket Players in the Presence of External
- Computational Definition of Strength and Weakness in the Presence of External Factors
- Learning Strength and Weakness Rules in the Presence of External FactorsExternal Factors
- Batting Analysis through CCA
- Bowling Analysis through CCA
- Experiments
- Batting Analysis - Steve Smith
- MINING STRENGTH AND WEAKNESS RULES OF CRICKET PLAYERS IN THE PRESENCE OF EXTERNAL FACTORS INFLUENCING THE GAME
In definition 6.1, when a stroke characteristic corresponds to an attack and includes any bowling function and any external characteristic. In Definition 6.1, when a stroke characteristic corresponds to a beat and includes any of the bowling characteristics and any external characteristic.
CCA1 (61.8%)
EXPERIMENTS
- Bowling Analysis - Kagiso Rabada
This outlier describes the order in which batsman Steve Smith batted in the match. This external feature depicts the session in which batsman Steve Smith batted in the match.
Web Application
Other rules: In windy weather, Rabada tends to bowl fast and away deliveries, and the batsmen play them in the third-man zone.
Summary
Visualization of Similar Players Based on their Strength and Weakness Rules
Visualization of Similar Players
- Visualization of Similar Batsmen
- Visualization of Similar Bowlers
For each batsman, the row/batting vector (I) and the column/bowling vectorGj corresponding to the first power rule are obtained. SNE Plot of WVBOWL t-SNE Plot of SVBOWL . abdur razzak - I'm leaving abdur rehman - soon adil rashid - moving. ajantha mendis - leg .. amit mishra - swing andre nel - swing .. ben hilfenhaus - leg ben stokes - move Away bhuvneshwar kumar - leg brett lee - swing .. chaminda vaas - move Away chanaka welegedara - swing triple - chrissga. - go away chris woakes - swing .. korey collymore - go away . dale steyn - swing daniel vettori - legs .. daren powell - leave . darren sammy - swing devendra bishoo - swing .. dhammika prasad - move away dilhara fernando - swing .. dilruwan perera - move away . doug bollinger - swing doug bracewell - move away . dwayne bravo - spin .. fidel edwards - moveAway graeme cremer - swing . graeme swann - moveAway harbhajan singh - leg .. iain obrien - moveAway imran tahir - swing . ex sodhi - fast - swing .. kane williamson - leg kemar roach - swing .. kyle abbott - move Lasith malinga - swing .. mahmudullah - leg makhaya ntini - swing mark craig - swing .. marlon samuels - fast . mashrafe mortaza - swing .. mat henry - leave matthew hoggard - foot .. mitchell marsh - move away mitchell santner - fast . mitchell starc - swing moeen ali - moveAway .. mohammad asif - slow mohammad hafeez - move away . mohammad sami - shake mohammad shami - leave . monty panesar - leaving .. munaf patel - leaving . muttiah muralitharan - move In nathan hauritz - leave . nathan lyon - moveIn .. nuwan kulasekara - leg nuwan pradeep - leg pat cummins - swing .. peter siddle - move away . pragyan ojha - move away rahat ali - leg .. rangana herath - move away ravi rampaul - spin .. ravichandran ashwin - move away ravindra jadeja - leave. roaston chase - swing rp singh - swing .. rubel hossain - leave ryan harris - leave. ryan sidebottom - leg saeed ajmal - swing .. shaminda eranga - leave shane shillingford - leave swing stuart wide - leave umesh yadav - leave. vernon philander - leave virender sehwag - moving .. yasir shah - fast . zaheer khan - swing zulfiqar babar - leg .. abdur razzak - swing abdur rehman - swing .. adil rashid - swing . ajantha mendis - swing amit mishra - short andre nel - leg .. angelo mathews - spinanil kumble - short ben hilfenhaus - full .. ben stokes - leg bhuvneshwar kumar - slow full chris gayle - short .. chris martin - spin chris tremlett - leg .. daniel vettori - short danish kaneria - short daren powell - short .. darren sammy - leg devendra bishoo - short . dhammika prasad - slow .. doug bollinger - spin doug bracewell - slow .. fidel edwards - full graeme cremer - short . graeme swann - short .. iain obrien - full imran tahir - swing ish sodhi - short .. ishant sharma - slow jacques kallis - spin .. james pattinson - spin jason holder - spin .. jeetan patel - short jerome taylor - leg .
Experiments
- Visualization of Similar Batsmen
- Visualization of Similar Bowlers
Each of these points is presented in the form of the batsman's name - bowling attributes, i.e. the batsman has shown weakness on the pitches that have the said bowling attribute. Each of these points is presented in the form of the bowler's name - bowling featurei, i.e. the bowler has shown weakness on the throws that have the said bowling feature.
Summary
Conclusions and Future Directions I n this chapter, we present the conclusions of the work carried out within the scope of this thesis
- Representing Cricket Text Commentary Data
- Mining Strength and Weakness Rules of Cricket Players
- Mining Temporal Changes in Strength and Weakness Rules of Cricket Playersof Cricket Players
- Mining Strength and Weakness Rules of Cricket Players in the Presence of External Factorsthe Presence of External Factors
- Visualization of Similar Players Based on their Strength and Weakness RulesWeakness Rules
- Future Research Directions
In Proceedings of the Sixth ACM International Conference on Web Search and Data Mining, 537âĂŞ546 (2013). In Proceedings of the Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.
Conference Papers
Conference Posters
Vedula, Learning Strengths and Weaknesses of Cricketers in the Presence of External Factors Affecting the Game.
Web Applications
He received his B.Tech degree in Computer Science and Engineering from Veer Surendra Sai University of Technology, Burla, India, in 2012 and his M.Tech degree in Computer Science and Engineering from the Indian Institute of Technology Guwahati, India, in July. 2015. Swarup began his PhD studies at the Department of Computer Science and Engineering, Indian Institute of Technology Guwahati, India, in July 2015 under the supervision of Drs.
Contact Information