Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

What Finance can learn from Dating Sites

3,541 views

Published on

Dating, as is often said, is a numbers game. And organizations such as Match.com, and Zoosk rely on very sophisticated technology as they sift through vast customer bases to create the most compatible couples. Specially, they rely on data to build the most nuanced portraits of their members that they can, so they can find the best matches. This is a business-critical activity for dating sites — the more successful the matching, the better revenues will be. One of the ways they do this is through graph databases. These differ from relational databases as they specialize in identifying the relationships between multiple data points. This means they can query and display connections between people, preferences and interests very quickly.

In this session you will see how in many ways dating sites are getting better performance and more value out of their data than financial institutions by using Neo4j.

  • Hello! High Quality And Affordable Essays For You. Starting at $4.99 per page - Check our website! https://vk.cc/82gJD2
       Reply 
    Are you sure you want to  Yes  No
    Your message goes here

What Finance can learn from Dating Sites

  1. 1. What  Finance  can  learn  from  Dating  Sites
 Max  De  Marzi
 GOTO  Chicago
  2. 2. About  Me • Max  De  Marzi  -­‐  Neo4j  Field  Engineer     • My  Blog:  http://maxdemarzi.com   • Find  me  on  Twitter:  @maxdemarzi   • Email  me:  maxdemarzi@gmail.com   • GitHub:  http://github.com/maxdemarzi
  3. 3. TLDR:
  4. 4. http://www.gartner.com/id=2081316 Consumer  Web  Giants  Depend  on  Five  Graphs Social
 Graph Mobile
 Graph Intent
 Graph Interest
 Graph Payment
 Graph
  5. 5. Friends  of  Friends  Graph   • Real  World  Basis   • Its  Weighted  (BFF  vs   Family)   • Awesome  or  Awkward The  Five  Graphs  of  Love Location  Graph   • Long  Distance  Sucks   • Where  are  the  single   people   • Where  should  we  meet Passion  Graph   • Shared  Interests   • Desired  Traits   • Long  vs  Short  Term Safety  Graph   • True  Identity   • Liers  and  Cheaters   • Balancing  Privacy SPAM  Graph   • Click  Bait   • Wanna  Cam?   • Professionals 1 2 3 4 5
  6. 6. The  Five  Graphs  of  Love 1Friends 2Passion 3Location 4Safety 5SPAM
  7. 7. Meet  Jeremy • Single   • Handsome   • Friendly   • etc Jeremy
  8. 8. Friends Johan Kerstin Allison Andreas Michael Madelene JeremyPeter
  9. 9. Bang  With  Friends  =>  Down
  10. 10. Better  Idea
  11. 11. Friends  of  Friends Joh Kers Allis An AdAndr Mich Madel JerePet
  12. 12. Deeper
  13. 13. Friends  of  Friends  of  Friends FOFOFs
  14. 14. MATCH  (:Person  {  name:“Dan”}  )  -­‐[:FRIENDS]-­‐>  (:Person  {  name:“Ann”}  )   FRIENDS Dan Ann Label Property Label Property Node Node Cypher  Query  Language
  15. 15. MATCH  (boss)-­‐[:MANAGES*0..3]-­‐>(sub),              (sub)-­‐[:MANAGES*1..3]-­‐>(report)   WHERE  boss.name  =  “John  Doe”   RETURN  sub.name  AS  Subordinate,  
    count(report)  AS  Total Express  Complex  Queries  Easily  with  Cypher Find  all  direct  reports  and  
 how  many  people  they  manage,  
 up  to  3  levels  down Cypher  QuerySQL  Query
  16. 16. The  Five  Graphs  of  Love 1Friends 2Passion 3Location 4Safety 5SPAM 2Passion
 Graph
  17. 17. Interests Jonathan :REPORTED_INTEREST
  18. 18. Likes
  19. 19. Traits
  20. 20. The  Five  Graphs  of  Love 1Friends 2Passion 3Location 4Safety 5SPAM 3Location
  21. 21. Location
  22. 22. Three  Dots  and  a  Dash
  23. 23. Recommend  Love Find  your  soulmate  in  the  graph     • Are  they  energetic?   • Do  they  like  dogs?   • Have  a  good  sense  of  humor?   • Neat  and  tidy,  but  not  crazy  about  it? What  are  the  Top  10  Potential  Mates  for  me   • that  are  in  the  same  location   • are  sexually  compatible   • have  traits  I  want     • want  traits  I  have
  24. 24. Cypher  Query:  Love  Recommendation
  25. 25. Love  Recommendation  Results
  26. 26. The  Five  Graphs  of  Love 1Friends 2Passion 3Location 4Safety 5SPAM 4Safety
  27. 27. Awkward  Graph :WANTS_TO_DATE :WANTS_TO_DATE JakePeterAndreas :WORKS_FOR:FRIENDS:FRIENDS :NO_DATE :NO_DATE :WANTS_TO_DATE :WANTS_TO_DATE Jennifer
  28. 28. Liars
  29. 29. Cheaters  Graph :WANTS_TO_DATE JakeLucyAndreas :LIKES :MARRIED:FRIENDS :NO_DATE :WANTS_TO_DATE :WANTS_TO_DATE Jennifer
  30. 30. Let’s  take  a  closer  look  at  Jonathan Jonathan
  31. 31. Follows
  32. 32. Tweets
  33. 33. Real  Interests :DEMONSTRATED_INTEREST Jonathan
  34. 34. The  Five  Graphs  of  Love 1Friends 2Passion 3Location 4Safety 5SPAM 5SPAM
  35. 35. Click  Bait
  36. 36. Cam  Girls/Boys
  37. 37. Professionals
  38. 38. Payment  Graph   • Fraud  detection   • Credit  risk  analysis   • Chargebacks Financial  Giants  Depend  on  Five  Graphs  As  Well   Asset  Graph   • Portfolio  analytics   • Risk  management   • Market/sentiment  analysis   • Compliance Customer  Graph   • Org  drill-­‐through   • Product   recommendations   • Mobile  payments Entitlement  Graph   • Identity  management   • Access  management   • Authorization Master  Data  Graph   • Enterprise  collaboration   • Corporate  hierarchy   • Data  governance 1 2 3 4 5
  39. 39. The  Five  Graphs  of  Finance 1Payment 2Customer 3Entitlement 4Asset 5Master  Data 1Payment
 Graph
  40. 40. Intuit  Payment  Graph Discover  latent  network  from  multiple  
 product  data  stores   • Uniquely  identify  entities  and  their   connections   • Connections  scored  by  volume  of  trade   Empower  business-­‐unit  teams  to  leverage  the   Intuit  Payment  Graph  to  build  applications   • Graph  to  be  available  for  real-­‐time  queries 1Payment
  41. 41. Consumer  Profile  Facets Identity   Name
 Address
 Phone
 Email Social   Facebook
 Yelp
 Twitter
 … Demographics   Age
 Gender
 … Business  Profile  Facets Identity   Name
 Address
 Phone
 Email
 Social   Facebook
 Yelp
 Twitter
 … Demographics   Category
 Revenue
 Employees
 … 1Payment Payment  Graph  Depends  on  the  Customer  Graph
  42. 42. 1Payment Capturing  C2B  and  B2B  Transactions BUSINESSBUSINESS CONSUMER June
 1  purchase
 $25.95 June
 3  purchases
 $650.25
  43. 43. PRODUCT   Name:  Zeta
 … PRODUCT   Name:  Payroll
 … COMPANY   Name:  Viva  LLC
 Zip:  94040
 … COMPANY   Name:  Beta  LLC
 Zip:  94043
 … COMPANY   Name:  Acme,  Inc.
 Zip:  95134
 … Relationship
 CUSTOMER   Transactions:  467
 Years:    3Relationship
 LICENSE
 Years:  8 Relationship
 CUSTOMER   Transactions:  125
 Years:    1 Relationship
 LICENSE
 Years:    3 #1:  Payment  Graph  Example
  44. 44. Streamlining  
 credit  card
 chargebacks Cardholder  calls
 card  issuer  to  dispute   transaction Cardholder  receives  
 credit  card  statement Card  issuer  returns   transaction  through   card  network Acquirer  resolves  chargeback  
 or  forwards  it  to  merchant Merchant  receives  chargeback  
 and  accepts  or  challenges  it Acquirer  forwards   representation  
 to  card  network Card  issuer  verifies   representation  and  
 credits  cardholder Network  verifies  and  
 forwards  representation  
 to  card  issuer #1:  Payment  Graph  Example 1Payment
  45. 45. 1Payment 2Customer 3Entitlement 4Asset 5Master  Data The  Five  Graphs  of  Finance 2Customer
 Graph
  46. 46. All  Companies  and  Customers  Are  Related 2Customer
  47. 47. The  Corporate   Hierarchy  is  
 really  a  graph 2Customer Corporate  Hierarchy  is  Really  a  Graph
  48. 48. Name   Windsor  Press,  Inc.                       Address   6  North  Third  St
             City   Hamburg   State   PA                     Zip   19526
                     Phone   610-­‐562-­‐2267 Name   The  Windsor  Press
                   Address   6  North  3rd  Street
 City   Hamburg   State   PA                     Zip   19526-­‐0465
                             Phone   610-­‐562-­‐2267 ID   002114902
                           Name   Windsor  Press,  Inc.   Address   6  N  3rd  St
             City   Hamburg   State   PA                     Zip   19526-­‐1502
                               Phone   610-­‐562-­‐2267 Both  of  the  records   above  map  to  the   same  record 2Customer Cleaning  and  Matching  for  360-­‐Degree  Master  View
  49. 49. Synthetic  Identities  and  Fraud  Rings 145  Hickory  Rd
 Pasadena,  CA 415  Hickory  St
 Pasadena,  CA 626-­‐407-­‐1234 626-­‐814-­‐6532 Quickly  see  which  customers  share  the   same  contact  information 2Customer
  50. 50. 3  fake  addresses  and  
 3  fake  phone  addresses
 can  create  9  fake  customers 2Customer Bank  Fraud  Using  False  Personas
  51. 51. High  Speed  Fraud  -­‐  1000  R/S http://maxdemarzi.com/2014/02/12/online-­‐payment-­‐risk-­‐management-­‐with-­‐neo4j/  
  52. 52. High  Speed  Fraud  -­‐  8000  R/S http://maxdemarzi.com/2014/02/27/neo4j-­‐at-­‐ludicrous-­‐speed/
  53. 53. High  Speed  Fraud  -­‐  28000  R/S http://maxdemarzi.com/2014/03/10/its-­‐over-­‐9000-­‐neo4j-­‐on-­‐websockets/
  54. 54. 1Payment 2Customer 3Entitlement 4Asset 5Master  Data The  Five  Graphs  of  Finance 3Entitlement
 Graph
  55. 55. http://blogs.gartner.com/ian-­‐glazer/2013/02/08/killing-­‐iam-­‐in-­‐order-­‐to-­‐save-­‐it/ Killing  IAM  in  Order  to  Save  It 3Entitlement
  56. 56. Access  Control
  57. 57. Permission  Resolution http://maxdemarzi.com/2013/03/18/permission-­‐resolution-­‐with-­‐neo4j-­‐part-­‐1/  
  58. 58. 1Payment 2Customer 3Entitlement 4Asset 5Master  Data The  Five  Graphs  of  Finance 4Asset
 Graph
  59. 59. Portfolio  Analytics Asset  Graph  Examples IT  Asset  Management Risk  Analysis 4Asset
  60. 60. Express  Complex  Relationship  Queries  Easily For  a  given  fund,  return  all  assets   that  are  made  up  of  other  assets,   ordered  by  the  total  number  of   included  assets Cypher  Query SQL  Query MATCH  (fund)-­‐[:INCLUDES*0..n]-­‐>(sub),              (sub)-­‐[:INCLUDES*1..n]-­‐>(asset)   WHERE  fund.ticker  =  “TRLGX”   RETURN  sub.ticker  AS  Asset_Group,  
            count(asset)  AS  Total   ORDER  BY  Total  DESC
  61. 61. The  Five  Graphs  of  Finance 1Payment 2Customer 3Entitlement 4Asset 5Master  Data 5Master  Data
 Graph
  62. 62. Research  Management
  63. 63. High  Frequency  Lies
  64. 64. Research  in  Context
  65. 65. Linked  Data Connect  to  the     Semantic  Web
  66. 66. Technology  for  Managing
 the  Five  Graphs  of  Finance 1Payment 2Customer 3Entitlement 4Asset 5Master  Data
  67. 67. Intercontinental  Exchange Social  network
 for  brokers 5Master
 Data
  68. 68. Hello  World  Recommendation
  69. 69. Hello  World  Recommendation
  70. 70. Movie  Data  Model
  71. 71. Cypher  Query:  Movie  Recommendation MATCH  (watched:Movie  {title:"Toy  Story”})  <-­‐[r1:RATED]-­‐  ()  -­‐[r2:RATED]-­‐>  (unseen:Movie)   WHERE  r1.rating  >  7  AND  r2.rating  >  7   AND  watched.genres  =  unseen.genres   AND  NOT(  (:Person  {username:”maxdemarzi"})  -­‐[:RATED|WATCHED]-­‐>  (unseen)  )   RETURN  unseen.title,  COUNT(*)   ORDER  BY  COUNT(*)  DESC   LIMIT  25 What  are  the  Top  25  Movies   • that  I  haven't  seen   • with  the  same  genres  as  Toy  Story     • given  high  ratings   • by  people  who  liked  Toy  Story
  72. 72. Relational  Databases  Can’t  Handle  Relationships  Well • Cannot  model  or  store  data  and  relationships   without  complexity   • Performance  degrades  with  number  &  levels  of   relationships,  and  database  size   • Query  complexity  grows  with  need  for  JOINs   • Adding  new  types  of    data  and  relationships   requires  schema  redesign,  increasing  time  to   market   …  making  traditional  databases  inappropriate  when   relationships  are  valuable  in  real-­‐time Slow  development
 Poor  performance
 Low  scalability
 Hard  to  maintain
  73. 73. NoSQL  Databases  Don’t  Handle  Relationships • No  data  structures  to  model  or  store   relationships   • No  query  constructs  to  support   relationships   • Relating  data  requires  “JOIN  logic”  in  the   application   • No  ACID  support  for  transactions   …  making  NoSQL  databases  inappropriate  when   relationships  are  valuable  in  real-­‐time
  74. 74. Real-­‐Time  Query  Performance
 Performance  must  hold  steady  with  scale Connectedness  and  Size  of  Data  Set Response  Time 0  to  2  hops
 0  to  3  degrees
 Thousands  of  connections Tens  to  hundreds  of  hops
 Thousands  of  degrees
 Billions    of  connections Relational  and
 Other  NoSQL
 Databases Neo4j Neo4j  is  
 1000x  faster
 Reduces  minutes  
 to  milliseconds
  75. 75. Re-­‐Imagine  Your  Data  as  a  Graph Neo4j  is  an  enterprise-­‐grade  graph   database  that  enables  you  to:   • Model  and  store  your  data  as  a   graph   • Query  relationships  with  ease   and  in  real-­‐time   • Seamlessly  evolve  applications   to  support  new  requirements  by  
 adding  new  kinds  of  data  and   relationships Agile  development
 High  performance
 Vertical  and  horizontal  scale
 Seamless  evolution
  76. 76. THANK  YOU

×