Kereskedelem | Elektronikus Kereskedelem » Supercharging Business Decisions with AI, Insights, Optimize and Personalize

Supercharging Business Decisions with AI: Ins ights , Optimize and Pers onalize How many Drivers do we need in December 2019 in San Francisco? How far do we plan? Tactic Strategic 4-6 weeks 12-18 months How do we know how many people are going to take Uber? Trip Forecasting Historical Trips Historical Trips data for a city Long-Term Forecasts Acquisition Spend Marketing (acquiring new riders/drivers) Events Big events in the city Upto 52 Weeks Time Series Model Used for year long budget planning Time-Series Forecas ting Algorithms Reference https://www.datasciencecom/blog/time-series-forecasting-machine-learning-differences Forecasting - Predictive Model Cohort Rider Driver Eater Month of joining First Trips Retention Rate Trips/Active User Trips Forecasting - Bayes ian Model It’s a probabilis tic graphical model that repres ents a s et of variables and their conditional dependencies via a graph (DAG). For example, a Bayes ian network could

repres ent the probabilis tic relations hips between dis eas es and s ymptoms . Given s ymptoms , the network can be us ed to compute the probabilities of the pres ence of various dis eas es . Reference https://en.wikipediaorg/wiki/Bayesian network#/media/File:SimpleBayesNetNodessvg Forecasting - Black Box Model Cohort Ensemble Bayesian Black Box Output Classical Backtesting Forecasting Models - Neural Net Trip transactions Marketing spending Cost Curve Model Trip Model Product Mix Model Cost Curve Model Trip Model Fare Model Ensemble Model Driver/Rider signup Retention Model Trip Model Incentive Model Trip Model UFP Up/Down Model Holiday/events Cost Model Net-Inflow Model Cost Model GB Service Fee Model User behavior Trips Net-Inflow Model Planning-as-a-Service Finance Modeling and Computation Platform Revenue Number of trips Number of Drivers needed But how do you balance the Market with Drivers & Riders Optimization Optimization

Process Financial Planning Budget Setting Regional Growth Leads Finance Ops Bi-annual Tool: Previous slides Output: monthly/weekly spend budget, by lever/PU Continuous On-demand Cross Lever Budgeting Tool: Cross Lever Optimizer (CLOe) Teams: Strategic + Regional Finance, Perf Marketing, Central Ops Output: updated weekly spend budget, by lever/PU Incentive Budget Spenders Ops Marketplace Marketing Paid Budget Referrals Budget Incentive Spend Paid Spend Referral Spend Tools: Finplan Output: Weekly EDI/ERI/UFP spend by city Tools: Mixed media model Output: Weekly marketing channel spend by city Tools: Web referrals Output: City referral structures Trip Forecasting & Optimization Historical Trips Long-Term Forecasts Historical Trips data for a city Acquisition Spend Upto 52 Weeks Time Series Model Used for year long budget planning Marketing (acquiring new riders/drivers) Short-Term Rolling Forecasts Driver Incentives (~ $1Bn) 1-12 Weeks Rider Promotions

(~ $500m) - Represents a spending lever Personalized Model Adjust budgets in spend levers to achieve trip targets Scenario Planning Scenario Generation Deviation from Forecast Weekly/monthly Incentive planning integrated with trip forecasting Which subset of users should we focus on to meet goal? Combining insights like these can help Uber adjust its budget in the short term. Optimization Model Overview User Level Model Paid + Organic Re fe rral Historical FTs Incentive spend Driver First time Drive r (FTD) by channe l Driver RR & TPA Drive r FT model (base d on channe l cost curve s) Paid + Organic Re fe rral First time Ride r (FTR) by channe l Trips = FT x RR x TPA Trips production function Legend Historical FTs Rider FT model (base d on channe l cost curve s) GB Trips = FT x RR x TPA Rider RR & TPA Promo Pe r trip me trics fore cast (Fare ) Model signal/input Cost curve Lever YYY Incentive spend Rider CLOe helps to optimize growth spend

across marketing, referrals and incentives. 16 Trip Model Detail (LSTM) y1 λ yn y2 π λ π λ y1 π λ yn y2 π λ π λ π FC FC FC FC FC FC LSTM 2 LSTM 2 LSTM 2 LSTM 2 LSTM 2 LSTM 2 I t+1 I t+2 I t’+1 I t+n LSTM 1 F t I t’+2 I t’+n LSTM 1 training F t’ n = 12 predict n = 12 LifeTime Value LTV is an estimate of LifeTime contribution of each user in order to drive efficiency in marketing, incentive spend and as a KPI to inform product improvements Model Overview Model Overview: We use Gradient Boosting Trees Mode l which consolidate s pre dictions of hundre ds of inde pe nde ntly traine d tre e s. Its an ite rative Mode l and Pre diction syste m Figure shows the use r le ve l GB Mode l pre diction. We are using the Gamma-Gamma BG/NBD mode l to pre dict the ne xt 2-ye ars of rolling gross bookings for e ach use r. Eng Platform: PySpark platform to proce ss Pe tabyte s of data Combine s Que ry, data frame s and

machine le arning Ability to acce ss data across Ube r’s data store s: Hive , HDFS, Cassandra, and S3. Apache Spark Ecosystem Spark SQL Streaming Machine Learning (MLlib) Graph Analytics (GraphX) Spark Core API R Python Scala Java Platform Finance Intelligence at Uber Forecasting Planning Budgeting Optimization Lifetime Value Model Orchestrator S Metrics Computation Management APIs Security Model Computation Service Forecasting Models Optimization Models LTV Models Finance Data Warehouse Data Pipelines Metrics Store Dashboards Analytics Scenario Management Service Data Platform Overview Financial Data Store (FDS) Future Looking forward to. Uber Freight Uber Health Drones (Food Delivery) Uber Elevate (Air Transportation) Autonomous Vehicles Facilitate better Transportation Proprietary and confidential 2018 Uber Technologies, Inc. All rights reserved No part of this docume nt may be re produce d or utilize d in any form or by

any me ans, e le ctronic or me chanical, including photocopying, re cording, or by any information storage or re trie val syste ms, without pe rmission in writing from Ube r. This docume nt is inte nde d only for the use of the individual or e ntity to whom it is addre sse d and contains information that is privile ge d, confide ntial or othe rwise e xe mpt from disclosure unde r applicable law. All re cipie nts of this docume nt are notifie d that the information containe d he re in include s proprie tary and confide ntial information of Ube r, and re cipie nt may not make use of, disse minate , or in any way disclose this docume nt or any of the e nclose d information to any pe rson othe r than e mploye e s of addre sse e to the e xte nt ne ce ssary for consultations with authorize d pe rsonne l of Ube r. Business Facts X Ride sharing cities, YUberEats cities $zz bn gross bookings (excluding Uber Eats) ??M+ active riders, ?M+ active drivers ??M+ trips/day Goal: Enable

intelligent models to make data-driven financial decisions faster and more accurately