FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950803838> ?p ?o ?g. }

• W2950803838 abstract "This paper describes a practical system for Multi Touch Attribution (MTA) foruse by a publisher of digital ads. We developed this system for JD.com, aneCommerce company, which is also a publisher of digital ads in China. Theapproach has two steps. The first step ('response modeling') fits a user-levelmodel for purchase of a product as a function of the user's exposure to ads.The second ('credit allocation') uses the fitted model to allocate theincremental part of the observed purchase due to advertising, to the ads theuser is exposed to over the previous T days. To implement step one, we train aRecurrent Neural Network (RNN) on user-level conversion and exposure data. TheRNN has the advantage of flexibly handling the sequential dependence in thedata in a semi-parametric way. The specific RNN formulation we implementcaptures the impact of advertising intensity, timing, competition, anduser-heterogeneity, which are known to be relevant to ad-response. To implementstep two, we compute Shapley Values, which have the advantage of havingaxiomatic foundations and satisfying fairness considerations. The specificformulation of the Shapley Value we implement respects incrementality byallocating the overall incremental improvement in conversion to the exposedads, while handling the sequence-dependence of exposures on the observedoutcomes. The system is under production at JD.com, and scales to handle thehigh dimensionality of the problem on the platform (attribution of the ordersof about 300M users, for roughly 160K brands, across 200+ ad-types, servedabout 80B ad-impressions over a typical 15-day period)." @default.
• W2950803838 created "2019-06-27" @default.
• W2950803838 creator A5004863713 @default.
• W2950803838 creator A5031508693 @default.
• W2950803838 creator A5041892633 @default.
• W2950803838 creator A5045169576 @default.
• W2950803838 creator A5088790197 @default.
• W2950803838 date "2019-02-01" @default.
• W2950803838 modified "2023-09-27" @default.
• W2950803838 title "Causally Driven Incremental Multi Touch Attribution Using a Recurrent Neural Network" @default.
• W2950803838 cites W1598796236 @default.
• W2950803838 cites W1819758051 @default.
• W2950803838 cites W1943882014 @default.
• W2950803838 cites W1991133427 @default.
• W2950803838 cites W2011256240 @default.
• W2950803838 cites W2027839911 @default.
• W2950803838 cites W2047995569 @default.
• W2950803838 cites W2064675550 @default.
• W2950803838 cites W2067619114 @default.
• W2950803838 cites W2092195356 @default.
• W2950803838 cites W2094149843 @default.
• W2950803838 cites W2123656122 @default.
• W2950803838 cites W2138116946 @default.
• W2950803838 cites W2143127637 @default.
• W2950803838 cites W2154668116 @default.
• W2950803838 cites W2155968480 @default.
• W2950803838 cites W2160592778 @default.
• W2950803838 cites W2162175235 @default.
• W2950803838 cites W2168802503 @default.
• W2950803838 cites W2175177695 @default.
• W2950803838 cites W2230300434 @default.
• W2950803838 cites W2259808229 @default.
• W2950803838 cites W2346728723 @default.
• W2950803838 cites W2462689321 @default.
• W2950803838 cites W2610600864 @default.
• W2950803838 cites W2808422032 @default.
• W2950803838 cites W2904729595 @default.
• W2950803838 cites W2964121744 @default.
• W2950803838 cites W2990479190 @default.
• W2950803838 cites W3007327886 @default.
• W2950803838 cites W3123047307 @default.
• W2950803838 cites W3124383212 @default.
• W2950803838 cites W3124604751 @default.
• W2950803838 cites W3124684770 @default.
• W2950803838 cites W3124849816 @default.
• W2950803838 cites W3125118956 @default.
• W2950803838 cites W1543099905 @default.
• W2950803838 hasPublicationYear "2019" @default.
• W2950803838 type Work @default.
• W2950803838 sameAs 2950803838 @default.
• W2950803838 citedByCount "4" @default.
• W2950803838 countsByYear W29508038382019 @default.
• W2950803838 countsByYear W29508038382020 @default.
• W2950803838 crossrefType "posted-content" @default.
• W2950803838 hasAuthorship W2950803838A5004863713 @default.
• W2950803838 hasAuthorship W2950803838A5031508693 @default.
• W2950803838 hasAuthorship W2950803838A5041892633 @default.
• W2950803838 hasAuthorship W2950803838A5045169576 @default.
• W2950803838 hasAuthorship W2950803838A5088790197 @default.
• W2950803838 hasConcept C111030470 @default.
• W2950803838 hasConcept C14036430 @default.
• W2950803838 hasConcept C143299363 @default.
• W2950803838 hasConcept C147168706 @default.
• W2950803838 hasConcept C154945302 @default.
• W2950803838 hasConcept C15744967 @default.
• W2950803838 hasConcept C41008148 @default.
• W2950803838 hasConcept C50644808 @default.
• W2950803838 hasConcept C77805123 @default.
• W2950803838 hasConcept C78458016 @default.
• W2950803838 hasConcept C86803240 @default.
• W2950803838 hasConceptScore W2950803838C111030470 @default.
• W2950803838 hasConceptScore W2950803838C14036430 @default.
• W2950803838 hasConceptScore W2950803838C143299363 @default.
• W2950803838 hasConceptScore W2950803838C147168706 @default.
• W2950803838 hasConceptScore W2950803838C154945302 @default.
• W2950803838 hasConceptScore W2950803838C15744967 @default.
• W2950803838 hasConceptScore W2950803838C41008148 @default.
• W2950803838 hasConceptScore W2950803838C50644808 @default.
• W2950803838 hasConceptScore W2950803838C77805123 @default.
• W2950803838 hasConceptScore W2950803838C78458016 @default.
• W2950803838 hasConceptScore W2950803838C86803240 @default.
• W2950803838 hasLocation W29508038381 @default.
• W2950803838 hasOpenAccess W2950803838 @default.
• W2950803838 hasPrimaryLocation W29508038381 @default.
• W2950803838 hasRelatedWork W1885141799 @default.
• W2950803838 hasRelatedWork W1988431231 @default.
• W2950803838 hasRelatedWork W2069000504 @default.
• W2950803838 hasRelatedWork W2186109963 @default.
• W2950803838 hasRelatedWork W2772691796 @default.
• W2950803838 hasRelatedWork W2781909030 @default.
• W2950803838 hasRelatedWork W2914271732 @default.
• W2950803838 hasRelatedWork W2915070458 @default.
• W2950803838 hasRelatedWork W2989700500 @default.
• W2950803838 hasRelatedWork W2998300237 @default.
• W2950803838 hasRelatedWork W3035029760 @default.
• W2950803838 hasRelatedWork W3098198982 @default.
• W2950803838 hasRelatedWork W3101314357 @default.
• W2950803838 hasRelatedWork W3111500217 @default.
• W2950803838 hasRelatedWork W3118697270 @default.
• W2950803838 hasRelatedWork W3120239804 @default.

• next