李仁杰/ Riot Games Head of Data Science

12,669 views

Published on

李仁傑,畢業於中國科學技術大學少年班, 美國 University of Rochester 博⼠。現為 Riot Games 數據科學總負責人。致力於通過開發創新的數據產品來提高用戶的體驗,甚至人類的生活。從⽆到有建立了Riot Games Data Science 部⻔, 帶領團隊從《英雄聯盟》的海量數據平台中用大數據分析大規模提升商業決策和運營效率, 並用機器學習算法開發了精細化運營, 精準分析預測, 個性化推薦, 自然語⾔處理等各種全球領先的數據產品。

Published in: Data & Analytics
1 Comment
21 Likes
Statistics
Notes
  • For Business Analytics Tools Online Training register at http://www.todaycourses.com
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here
No Downloads
Views
Total views
12,669
On SlideShare
0
From Embeds
0
Number of Embeds
9,648
Actions
Shares
0
Downloads
288
Comments
1
Likes
21
Embeds 0
No embeds

No notes for slide

李仁杰/ Riot Games Head of Data Science

  1. 1. DATA SCIENCE RENJIE LI AT RIOT GAMES 07.16.2016
  2. 2. 2009 LAUNCH TEAM ORIENTED ONLINE GAME 120+ CHAMPS LIVE PLAYERS VS. LIVE PLAYERS
  3. 3. BASKETBALL LEAGUE OF LEGENDS
  4. 4. Players & Data W O R L D W I D E Statistics released Jan 2014 67+ million monthly active players 500+ billion data points per day 26 petabytes data collected since beta
  5. 5. DATA SCIENCE AT RIOT GAMES EMPOWER RIOTERS TO MAKE BETTER DATA POWERED PRODUCTS TEAM MISSION
  6. 6. DATA SCIENCE AT RIOT GAMES PLAYER FOCUSED DATA INFORMED, NOT DATA DRIVEN TEAM PHILOSOPHIES
  7. 7. DATA DRIVEN INFORMED DECISION MAKING DATA DRIVEN DATA INFORMED
  8. 8. DATA SCIENCE AT RIOT GAMES Data Science
  9. 9. DATA SCIENCE AT RIOT GAMES Data Science RiskMarketing Ecommerce
  10. 10. DATA SCIENCE AT RIOT GAMES Data Science Social Play RiskMarketing Ecommerce
  11. 11. DATA SCIENCE AT RIOT GAMES Data Science Social Play Risk Match Making Marketing Ecommerce
  12. 12. DATA SCIENCE AT RIOT GAMES Data Science Social Play Risk Match Making Player Onboarding Marketing Ecommerce
  13. 13. DATA SCIENCE AT RIOT GAMES Data Science Social Play AI Risk R & D Match Making Eco System Player Support Player Onboarding Game Balance Design Marketing Ecommerce
  14. 14. CASE STUDY #1 UNDERSTANDING PLAYERS’ IN-GAME BEHAVIOR
  15. 15. UNDERSTANDING PLAYERS’ IN-GAME BEHAVIOR CHAMPION PLAY PATTERNS
  16. 16. UNDERSTANDING PLAYERS’ IN-GAME BEHAVIOR CHAMPION PLAY PATTERNS
  17. 17. UNDERSTANDING PLAYERS’ IN-GAME BEHAVIOR CHAMPION PLAY PATTERNS TYPE A CHAMPION TYPE B CHAMPION TYPE C CHAMPION TYPE E CHAMPION TYPE F CHAMPION TYPE G CHAMPION TYPE D CHAMPION 7 KEY TYPES OF CHAMPIONS OUR DESIGN PHILOSOPHY ALIGNS WITH PLAYERS PLAY STYLES
  18. 18. UNDERSTANDING PLAYERS’ IN-GAME BEHAVIOR 0%   10%   20%   30%   40%   50%   60%   Type  A   Champion   Type  B   Champion   Type  C   Champion   Type  D   Champion   Type  E   Champion   Type  F   Champion   Type  G   Champion   Player  1   Player  2   PLAYER SEGMENTATION BASED ON HOW THEY PLAY EACH TYPE OF CHAMPION
  19. 19. UNDERSTANDING PLAYERS’ IN-GAME BEHAVIOR PLAYER SEGMENTATION BASED ON HOW THEY PLAY EACH TYPE OF CHAMPION 9 DIFFERENT CHAMPION PLAY BEHAVIOR
  20. 20. UNDERSTANDING PLAYERS’ IN-GAME BEHAVIOR Type  A   Type  B   Type  C   Type  D   Type  E   Type  F   Type  G   IDEAL VS. CURRENT DISTRIBUTION OF CHAMPION CLUSTERS CURRENT IDEAL OVERSERVEDUNDERREPRESENTED
  21. 21. CASE STUDY #1 TAKEAWAYS CERTAIN CHAMPION ARCHETYPES MAY BE UNDERREPRESENTED OR OVERSERVED HOW DIFFERENTLY PLAYERS PLAY OUR CHAMPIONS WHETHER OUR DESIGN PHILOSOPHY ALIGNS WITH PLAYERS’ PLAY STYLES CHAMPION PLAY PATTERN MODEL HELPS US BETTER UNDERSTAND:
  22. 22. CASE STUDY #2 UNDERSTANDING PLAYER ENGAGEMENT
  23. 23. UNDERSTANDING PLAYER ENGAGEMENT week 1 week 2 week 3 week 4 week 5 week 6 week 7 week 8 week 9 week 10 week 11 week 12 week 13 week 14 week 15 week 16 week 17 week 18 week 19 week 20 week 21 week 22 week 23 week 24 Actual   TOTALHOURSPLAYED EVENTB EVENTC EVENTD EVENTE EVENTA
  24. 24. UNDERSTANDING PLAYER ENGAGEMENT week 1 week 2 week 3 week 4 week 5 week 6 week 7 week 8 week 9 week 10 week 11 week 12 week 13 week 14 week 15 week 16 week 17 week 18 week 19 week 20 week 21 week 22 week 23 week 24 Predicted   TOTALHOURSPLAYED EVENTB EVENTC EVENTD EVENTE EVENTA
  25. 25. UNDERSTANDING PLAYER ENGAGEMENT week 1 week 2 week 3 week 4 week 5 week 6 week 7 week 8 week 9 week 10 week 11 week 12 week 13 week 14 week 15 week 16 week 17 week 18 week 19 week 20 week 21 week 22 week 23 week 24 Predicted   Actual   TOTALHOURSPLAYED EVENTB EVENTC EVENTD EVENTE EVENTA
  26. 26. CASE STUDY #2 TAKEAWAYS PROVIDE INSIGHTS FOR STRATEGIC DECISION MAKING AND PLANNING QUANTIFY EFFECT OF EVENTS AND CHANGES IN THE GAME FIND AND UNDERSTAND IMPORTANT FACTORS THAT PLAYERS ARE INTERESTED IN WEEKLY ENGAGEMENT PREDICTION MODEL HELPS US :
  27. 27. Production Predictive Diagnostic Descriptive
  28. 28. CASE STUDY #3 PERSONALIZED RECOMMENDATION
  29. 29. PERSONALIZED RECOMMENDATION MANY ENTERTAINMENT SERVICES AND ECOMMERCE COMPANIES HAVE MADE RECOMMENDER SYSTEM A PROMINENT PART OF THEIR WEBSITES GOOD RECOMMENDATIONS ADD ANOTHER DIMENSION TO THE USER EXPERIENCE AND BOOST CONTENT ENGAGEMENT
  30. 30. PERSONALIZED RECOMMENDATION GOOGLE NEWS: RECOMMENDATIONS GENERATE 38% MORE CLICKTHROUGH NETFLIX: 66% OF THE MOVIES WATCHED ARE RECOMMENDED AMAZON: 35% SALES ARE FROM RECOMMENDATIONS
  31. 31. PERSONALIZED RECOMMENDATION SIMILAR RECOMMENDATION The recommender system is a predictive model that generates recommendations based on similarities between users and items as well as user-item interaction.
  32. 32. PERSONALIZED RECOMMENDATION
  33. 33. PERSONALIZED RECOMMENDATION
  34. 34. PERSONALIZED RECOMMENDATION
  35. 35. CASE STUDY #4 INSTANT FEEDBACK SYSTEM
  36. 36. TRIBUNAL
  37. 37. NATURAL LANGUAGE PROCESSING
  38. 38. NATURAL LANGUAGE PROCESSING
  39. 39. INSTANT FEEDBACK SYSTEM TRIBUNAL NATURAL LANGUAGE PROCESSING+
  40. 40. INSTANT FEEDBACK SYSTEM SUPPORTS 14 DIFFERENT LANGUAGES INCREASES CHATLOG COVERAGE FROM 10% TO 100% DECREASES DETECTION TIME FROM WEEKS to 15 MINS
  41. 41. DATA SCIENCE AT RIOT GAMES PLAYER FOCUSED DATA INFORMED, NOT DATA DRIVEN TEAM PHILOSOPHIES EMPOWER RIOTERS TO MAKE BETTER DATA POWERED PRODUCTS TEAM MISSION
  42. 42. DATA SCIENCE Renjie Li rli@riotgames.com THANK YOU! TW OFFICE Wayne Lee Julia Su walee@riotgames.com jsu@riotgames.com
  43. 43. QUESTIONS?

×