Data Science Projects @ Runnr

Download as pptx, pdf

1 like180 views

This document summarizes a data science talk that covered the following topics: 1) An introduction to data science, including what it is, where data comes from, and who uses it. Data scientists use skills in math, statistics, programming and domains to model data, experiment, validate models and deploy solutions. 2) Examples of data science applications at Roadrunnr to minimize delivery times and costs while improving reliability, such as optimizing multi-drop grouping logic for deliveries and predicting demand patterns. 3) Future data science projects at Roadrunnr, including surge pricing, wait time prediction, carrier selection, driver fraud detection, and product size estimation with images.

Data & Analytics

Data Science Talk
AMOL SAHASRABUDHE
ANKIT JAIN

WHY THIS PRESENTATION
DEMOCRATIZE
BRAINSTORMING
2

? Why and What of Data Science
? Data Science @Roadrunnr
? Multi Drop Logic
? Demand Prediction
? Future Work
Agenda
3

HOUSE OF CARDS
WHO IS A DATA SCIENTIST
8

HOUSE OF CARDS
WHAT DO THEY DO
IDEA EXPERIMENT VALIDATE SPEC DEPLOY
MODELING
SKILLS : Math, Statistics, Programming and Domain Knowledge
10

HOUSE OF CARDS
DATA SCIENCE@ROADRUNNR
11

Use data as a raw input to develop products to:
? Minimize Estimated Time of Arrival (ETA)
? Improve Reliability
at minimal Costs
VALUES
12

Multi Drop Clubbing Logic
? Improve the multi drop grouping logic to:
? Reduce number of touchpoints/order (5-15%)
? Reduce distance travelled per touch point
? Increase driver satisfaction
13

Existing Multi Drop Logic
HUB
Red Cluster can be done away with in this scenario
14

Ideal Clubbing
HUB
DB should be able to cover points on his way
15

How to Achieve Ideal Clubbing?
HUB
D1 < D2
Join two points which have minimum distance between them
? Joining two points
16

Joining Two Groups
? Join on the basis of shortest distances and not center distances to
achieve ideal clubbing
HUB
? S1, S2 shortest distances
? C1, C2 center distances
17

Delhi West Hub Orders (Existing Clubbing)
18

Algorithm Limitations
? Sub optimal solution due to accuracy vs complexity trade off
? Performance limited by accuracy of Lat,Long of drop points
? Not optimized for size and weight of shipments
? Routing not included as a part of this version (V2)
? Considers only bike as a carrier (V2)
20

Demand Prediction (Impact)
?30% stock-outs during peak hours
?70% demand during peak hours
?Supply Planning
?Driver Placement
?Surge Pricing
22

Demand Prediction(Actual Demand)
Koramangala
High Variance in hourly demand
23

Demand Prediction(Actual and Predicted)
Koramangala
25

Demand Prediction(drill down)
Koramangala 8PM
Predictability in number of orders in this hour
26

Demand Prediction(Model Failure)
Electronic City
Demand is too erratic to be captured by modeling
Bad predictions
27

Demand Prediction(Future Work)
?Category specific demand
?Prediction for smaller clusters rather than localities
?Prediction over smaller intervals
?Stock-out incorporation
28

Data Science Future Projects
?Surge pricing
?Wait time prediction for food orders (deployment pending)
?Carrier Selection (Extension of multi drop logic)
?Driver Fraud detection
?Product size estimation using image
29

Walkalytics provides a reachability analysis tool using isochrones to measure walking times from any starting point in a neighborhood. This system allows businesses to optimize service locations, improve customer relationships, and effectively target audiences, making it useful for various sectors like public transport and retail. The document outlines the technology's advantages, such as worldwide coverage and the ability to integrate custom data.

Life LessonsAnkit Jain

Data analytics in fraud detection and customer feedbackAnkit Jain

This document discusses how data analytics can be used for fraud detection and analyzing customer feedback in ecommerce. It outlines common types of ecommerce frauds committed by buyers and sellers. It then describes how machine learning can be used to identify fraud buyers based on labeled transaction data and generated features. Customer feedback is also discussed, highlighting metrics like net promoter score and how natural language processing and bag of words models can analyze sentiment and pain points from reviews.

Data analytics workshop @IIIT BangaloreAnkit Jain

This document discusses data analytics and the differences between data analysts and data scientists. It provides examples of where data science is used in everyday products from Google, Facebook, Uber, Amazon and others. A case study is presented on how Yammer investigated a drop in user engagement through data analysis. They found the major engagement drop was for mobile users, and further exploration showed the issue was likely related to long-time user retention on the mobile app.

Data Science in EcommerceAnkit Jain

Data science is increasingly becoming vital in the e-commerce and fashion industries, offering objective insights into customer preferences through data analysis. Key applications include inventory management, personalized marketing, fraud detection, and optimizing customer interactions. The organizational structure typically involves data scientists embedded in various functions, collaborating with product managers and teams to enhance decision-making and consumer experience.

Advanced regression and model selectionAnkit Jain

The document discusses various machine learning algorithms and model selection techniques, highlighting the advantages and disadvantages of each method, including Naive Bayes, logistic regression, decision trees, and SVM. It also covers advanced regression issues and regularization techniques, detailing ridge and lasso regression's pros and cons along with their applications in gene selection problems. Lastly, it introduces additional methods such as Poisson regression and piecewise linear regression for specific types of data.

Operationalizing Machine Learning Using GPU-accelerated, In-database AnalyticsKinetica

The document discusses the advantages of GPU-accelerated, in-database analytics for operationalizing machine learning, emphasizing improvements in big data processing, AI/ML lifecycle management, and real-time analytics. It highlights the inefficiencies of traditional AI processes and presents Kinetica as a streamlined solution with integrated CPU/GPU support for various applications, including geospatial data analysis and real-time streaming. Case studies demonstrate significant performance enhancements for organizations like the US Army, showcasing how Kinetica can enable rapid geospatial and analytics capabilities while reducing costs and resource usage.

Hybrid Transactional/Analytics Processing with Spark and IMDGsAli Hodroj

This document discusses hybrid transactional/analytical processing (HTAP) with Apache Spark and in-memory data grids. It begins by introducing the speaker and GigaSpaces. It then discusses how modern applications require both online transaction processing and real-time operational intelligence. The document presents examples from retail and IoT and the goals of minimizing latency while maximizing data analytics locality. It provides an overview of in-memory computing options and describes how GigaSpaces uses an in-memory data grid combined with Spark to achieve HTAP. The document includes deployment diagrams and discusses data grid RDDs and pushing predicates to the data grid. It describes how this was productized as InsightEdge and provides additional innovations and reference architectures.

Uzair's CVuzair rasheed

Uzair Rasheed is seeking a position in a reputable telecommunications organization where he can apply his technical skills and research-based practical knowledge of emerging technologies. He has a B.Sc. in Telecommunication Engineering and over 5 years of experience in drive testing, site installation and maintenance, and troubleshooting for 2G, 3G, and 4G networks. His technical skills include IBS solutions, network planning and design, drive test KPI analysis, site surveys, configuration of networking equipment, and software like TEMS and LMT.

Performance Models for Apache AccumuloSqrrl

This document discusses performance optimization of Apache Accumulo, a distributed key-value store. It describes modeling Accumulo's bulk ingest process to identify bottlenecks, such as disk utilization during the reduce phase. Optimization efforts included improving data serialization to speed sorting, avoiding premature data expansion, and leveraging compression. These techniques achieved a 6x speedup. Current Accumulo performance projects include optimizing metadata operations and write-ahead log performance.

Webinar: SQL for Machine Data?Crate.io

The document presents a discussion on the significance of machine data management and the evolution of SQL to accommodate it, emphasizing its growing role in the Internet of Things (IoT) by 2020. Several customer use cases highlight the advantages of CrateDB over traditional databases like MySQL and Elasticsearch, showcasing improvements in performance, cost-effectiveness, and real-time data processing. The document concludes that new DBMS architectures are essential for effectively managing the increasingly complex and voluminous machine data produced in modern business environments.

Accumulo Summit 2015: Performance Models for Apache Accumulo: The Heavy Tail ...Accumulo Summit

The document discusses performance optimization techniques for Apache Accumulo, focusing on modeling distributed application performance and improving bulk loading speeds. It highlights the complexities of shared-nothing architectures and presents a performance analysis cycle aimed at enhancing throughput. Key strategies include refining code, simulating individual components, and leveraging serialization improvements to achieve significant speedups in data processing.

Research Methodology Presentation - Research in Supply Chain Digital TwinsArwa Abougharib

This document presents a theoretical framework for optimizing cold supply chains through the implementation of digital twin technology, integrating IoT, blockchain, and AI to enhance efficiency and product quality. It identifies challenges faced by supply chains, explores technological solutions, and outlines a research methodology to assess e-readiness and develop strategic frameworks. The findings suggest that digital twins can improve logistics operations and decision-making within cold supply chains, while also addressing issues related to data accuracy and reliability.

Large-scaled telematics analyticsDataWorks Summit

The document discusses the development and implementation of a large-scale telematics analytics system at Uber, outlining its engineering pipeline and data processing methods. It includes details on data sources, processing frameworks, and challenges faced, such as handling large volumes of trips and managing engineering problems like out-of-memory errors. Additionally, the authors emphasize the importance of modular design and data reuse in their analytical processes.

Integrating Semantic Web with the Real World - A Journey between Two Cities ...Juan Sequeda

The document explores the commercialization of semantic web technology to improve data integration and business intelligence. It discusses challenges and historical context, emphasizing the importance of knowledge engineering and effective mappings between relational databases and the semantic web. Strategies for overcoming barriers, such as optimizing query performance and the need for specialized roles like knowledge engineers, are also highlighted.

Stinger Initiative: Leveraging Hive & Yarn for High-Performance/Interactive Q...Caserta

The document discusses the Stinger initiative aimed at enhancing Hive's performance through the Tez execution engine, which moves beyond MapReduce to deliver improved query interactivity and speed. It highlights features such as optimized SQL coverage, in-memory operations, and an intelligent optimizer to streamline analytics on Hadoop. Furthermore, the Tez service minimizes overhead by eliminating task-launch latencies, making it suitable for short, interactive queries.

Hardware Accelerated Machine Learning Solution for Detecting Fraud and Money ...TigerGraph

The document discusses a hardware-accelerated machine learning solution utilizing FPGA technology for detecting fraud and money laundering. It highlights the advantages of graph analytics and machine learning in uncovering complex patterns and relationships within data, specifically for fraud detection use cases. The presentation includes technical details about the TigerGraph platform and its integration with Xilinx FPGAs to enhance performance in real-time analytics.

Scylla Summit 2016: ScyllaDB, Present and FutureScyllaDB

The document outlines ScyllaDB's commitments focusing on high throughput, low latency, compatibility with Apache Cassandra, and workload conditioning. It details the roadmap for short and long-term features, including improvements in performance, support for various workloads, and innovations like multi-tenancy and a new storage format. The emphasis is on reducing operational costs and enhancing compatibility while simplifying management through modern approaches like REST APIs.

PLNOG 3: John Evans - Best Practices in Network PlanningPROIDEA

This document discusses best practices in network planning. It recommends starting with a simple approach using link statistics and estimation to determine traffic demand matrices, then verifying predictions against reality. If needed, more data from NetFlow or MPLS statistics can be added incrementally. Simulation is used to analyze working and failure scenarios to determine capacity requirements while meeting SLAs with a modest overprovisioning factor of around 10%.

CEP - simplified streaming architecture - Strata Singapore 2016Mathieu Dumoulin

The document discusses Complex Event Processing (CEP) as a method for analyzing streaming event data to identify and respond to meaningful events in real-time. It emphasizes the need for modern architectures that support CEP and highlights the advantages of integrating rule engines with streaming technologies like Kafka. The presentation also addresses the scalability of these systems and how they can be effectively utilized in applications such as IoT and smart city traffic management.

Big Data 2.0 - Milwaukee Big Data User Group Presentation NVISIA

The document provides a technology overview for database management systems (DBMS) suitable for reporting needs over the next 3-5 years, evaluating various technologies such as RDBMS, NoSQL, and in-memory databases. Key evaluation criteria include scalability, functionality, cost, and drop-in compatibility, with a focus on the performance characteristics of each technology type. The summary emphasizes the importance of considering both traditional and newer database technologies due to falling hardware prices and niche technology availability.

OPTIMIZING THE TICK STACKInfluxData

The document discusses optimizing the Tick Stack for InfluxDB, focusing on query performance and effective schema design. It highlights best practices such as proper use of tags versus fields, correcting data precision, and ensuring efficient organization of databases. Case studies illustrate advantages and disadvantages of tag and field usage for sensor data, while emphasizing the importance of using tools like Telegraf for data collection.

Customer Story: Elastic Stack? ??? ?? ??? ?? ?? ???Elasticsearch

Builton emphasizes the necessity of e-commerce big-data analysis as the industry grows, providing insights into consumer behavior, product sales strategies, and competitive analysis. The company transitioned to using Elasticsearch to improve performance, scalability, and response times for data processing and visualization compared to their previous RDBMS system. Future plans include enhancements such as virtualization, monitoring, and infographics, leveraging the capabilities of Elasticsearch.

Approximate shortest distance computingLeMeniz Infotech

The document discusses a proposed query-dependent local landmark scheme for accurately estimating shortest distances in large graphs, addressing challenges posed by traditional global landmark methods. It outlines inefficiencies in classical algorithms and offers optimization techniques for graph compression and online queries. Additionally, the document provides hardware and software requirements for implementing the proposed system.

Bandwidth distributed denial of service attacks and defensesLeMeniz Infotech

The document presents a query-dependent local landmark scheme aimed at improving the accuracy of distance estimation in large graphs. It identifies local landmarks for query nodes and introduces efficient indexing and retrieval techniques to enhance query performance. Furthermore, it addresses the challenges of handling massive graph data and specifies hardware and software requirements for implementation.

Real-Time Streaming: Move IMS Data to Your Cloud Data WarehousePrecisely

The document discusses the importance and challenges of moving IMS data to cloud data warehouses (CDWs), highlighting best practices for effective integration. It addresses the significant role of IMS within major corporations, particularly in relation to transaction data, while outlining common issues and strategies for improving data transfer processes. Additionally, it emphasizes the need for a structured and incremental approach to minimize data overload and maximize efficiency in streaming performance.

Data Scotland - Migrating Mapping Dataflows by Johan Kangasniemi.pdfkangasniemi

The document presents a talk by Johan Kangasniemi on migrating mapping dataflows from Azure Synapse Analytics to Spark notebooks, focusing on the differences, migration challenges, and costs involved. It covers three main migration strategies: manual migration, syntactical parsing, and LLM-based migration, each with its own implications depending on the complexity of the dataflows. The talk highlights that despite the various options available, manual intervention is typically necessary during the migration process.

An Enterprise Architect's View of MongoDBMongoDB

This document discusses how MongoDB can help enterprises meet modern data and application requirements. It outlines the many new technologies and demands placing pressure on enterprises, including big data, mobile, cloud computing, and more. Traditional databases struggle to meet these new demands due to limitations like rigid schemas and difficulty scaling. MongoDB provides capabilities like dynamic schemas, high performance at scale through horizontal scaling, and low total cost of ownership. The document examines how MongoDB has been successfully used by enterprises for use cases like operational data stores and as an enterprise data service to break down silos.

美国毕业证范本中华盛顿大学学位证书颁奥鲍学生卡购买Taqyea

1:1原版中华盛顿大学毕业证+CWU成绩单【Q微:1954 292 140】鉴于此，CWUdiploma中华盛顿大学挂科处理解决方案CWU毕业证成绩单专业服务学历认证【Q微:1954 292 140】办理教育部学历认证，留学回国证明，中华盛顿大学毕业证、中华盛顿大学成绩单、中华盛顿大学文凭（留信学历认证+永久存档查询）办理本科+硕士+博士毕业证成绩单学历认证，我们一直是留学生的首选，质量行业第一，诚信可靠。【中华盛顿大学成绩单一站式办理专业技术完美呈现Central Washington University Transcripts】购买日韩成绩单、英国大学成绩单、美国大学成绩单、澳洲大学成绩单、加拿大大学成绩单（q微1954292140）新加坡大学成绩单、新西兰大学成绩单、爱尔兰成绩单、西班牙成绩单、德国成绩单。成绩单的意义主要体现在证明学习能力、评估学术背景、展示综合素质、提高录取率，以及是作为留信认证申请材料的一部分。中华盛顿大学成绩单能够体现您的的学习能力，包括中华盛顿大学课程成绩、专业能力、研究能力。（q微1954292140）具体来说，成绩报告单通常包含学生的学习技能与习惯、各科成绩以及老师评语等部分，因此，成绩单不仅是学生学术能力的证明，也是评估学生是否适合某个教育项目的重要依据！【主营项目】一.毕业证【q微1954292140】成绩单、使馆认证、教育部认证、雅思托福成绩单、学生卡等！二.真实使馆公证(即留学回国人员证明,不成功不收费) 三.真实教育部学历学位认证（教育部存档！教育部留服网站永久可查）四.办理各国各大学文凭(一对一专业服务,可全程监控跟踪进度)

UPS and Big Data intro to Business Analytics.pptxsanjum5582

More Related Content

Similar to Data Science Projects @ Runnr (20)

Uzair's CVuzair rasheed

Performance Models for Apache AccumuloSqrrl

Webinar: SQL for Machine Data?Crate.io

Accumulo Summit 2015: Performance Models for Apache Accumulo: The Heavy Tail ...Accumulo Summit

Research Methodology Presentation - Research in Supply Chain Digital TwinsArwa Abougharib

Large-scaled telematics analyticsDataWorks Summit

Integrating Semantic Web with the Real World - A Journey between Two Cities ...Juan Sequeda

Stinger Initiative: Leveraging Hive & Yarn for High-Performance/Interactive Q...Caserta

Hardware Accelerated Machine Learning Solution for Detecting Fraud and Money ...TigerGraph

Scylla Summit 2016: ScyllaDB, Present and FutureScyllaDB

PLNOG 3: John Evans - Best Practices in Network PlanningPROIDEA

CEP - simplified streaming architecture - Strata Singapore 2016Mathieu Dumoulin

Big Data 2.0 - Milwaukee Big Data User Group Presentation NVISIA

OPTIMIZING THE TICK STACKInfluxData

Customer Story: Elastic Stack? ??? ?? ??? ?? ?? ???Elasticsearch

Approximate shortest distance computingLeMeniz Infotech

Bandwidth distributed denial of service attacks and defensesLeMeniz Infotech

Real-Time Streaming: Move IMS Data to Your Cloud Data WarehousePrecisely

Data Scotland - Migrating Mapping Dataflows by Johan Kangasniemi.pdfkangasniemi

An Enterprise Architect's View of MongoDBMongoDB

Uzair's CVuzair rasheed

Performance Models for Apache AccumuloSqrrl

Webinar: SQL for Machine Data?Crate.io

Accumulo Summit 2015: Performance Models for Apache Accumulo: The Heavy Tail ...Accumulo Summit

Research Methodology Presentation - Research in Supply Chain Digital TwinsArwa Abougharib

Large-scaled telematics analyticsDataWorks Summit

Integrating Semantic Web with the Real World - A Journey between Two Cities ...Juan Sequeda

Stinger Initiative: Leveraging Hive & Yarn for High-Performance/Interactive Q...Caserta

Hardware Accelerated Machine Learning Solution for Detecting Fraud and Money ...TigerGraph

Scylla Summit 2016: ScyllaDB, Present and FutureScyllaDB

PLNOG 3: John Evans - Best Practices in Network PlanningPROIDEA

CEP - simplified streaming architecture - Strata Singapore 2016Mathieu Dumoulin

Big Data 2.0 - Milwaukee Big Data User Group Presentation NVISIA

OPTIMIZING THE TICK STACKInfluxData

Customer Story: Elastic Stack? ??? ?? ??? ?? ?? ???Elasticsearch

Approximate shortest distance computingLeMeniz Infotech

Bandwidth distributed denial of service attacks and defensesLeMeniz Infotech

Real-Time Streaming: Move IMS Data to Your Cloud Data WarehousePrecisely

Data Scotland - Migrating Mapping Dataflows by Johan Kangasniemi.pdfkangasniemi

An Enterprise Architect's View of MongoDBMongoDB

Recently uploaded (20)

美国毕业证范本中华盛顿大学学位证书颁奥鲍学生卡购买Taqyea

UPS and Big Data intro to Business Analytics.pptxsanjum5582

presentation4.pdf Intro to mcmc methodssSergeyTsygankov6

Microsoft Power BI - Advanced Certificate for Business Intelligence using Pow...Prasenjit Debnath

NASA ESE Study Results v4 05.29.2020.pptxCiroAlejandroCamacho

NVIDIA Triton Inference Server, a game-changing platform for deploying AI mod...Tamanna36

25 items quiz for practical research 1 in grade 11leamaydayaganon81

Indigo dyeing Presentation (2).pptx as dyeshreeroop1335

Data Visualisation in data science for studentsconfidenceascend

MRI Pulse Sequence in radiology physics.pptxBelaynehBishaw

Indigo_Airlines_Strategy_Presentation.pptxmukeshpurohit991

Camuflaje Tipos Características Militar 2025.ppte58650738

最新版美国威斯康星大学河城分校毕业证（鲍奥搁贵毕业证书）原版定制taqyea

2025原版威斯康星大学河城分校毕业证书pdf电子版【q薇1954292140】美国毕业证办理UWRF威斯康星大学河城分校毕业证书多少钱？【q薇1954292140】海外各大学Diploma版本，因为疫情学校推迟发放证书、证书原件丢失补办、没有正常毕业未能认证学历面临就业提供解决办法。当遭遇挂科、旷课导致无法修满学分，或者直接被学校退学，最后无法毕业拿不到毕业证。此时的你一定手足无措，因为留学一场，没有获得毕业证以及学历证明肯定是无法给自己和父母一个交代的。【复刻威斯康星大学河城分校成绩单信封,Buy University of Wisconsin-River Falls Transcripts】购买日韩成绩单、英国大学成绩单、美国大学成绩单、澳洲大学成绩单、加拿大大学成绩单（q微1954292140）新加坡大学成绩单、新西兰大学成绩单、爱尔兰成绩单、西班牙成绩单、德国成绩单。成绩单的意义主要体现在证明学习能力、评估学术背景、展示综合素质、提高录取率，以及是作为留信认证申请材料的一部分。威斯康星大学河城分校成绩单能够体现您的的学习能力，包括威斯康星大学河城分校课程成绩、专业能力、研究能力。（q微1954292140）具体来说，成绩报告单通常包含学生的学习技能与习惯、各科成绩以及老师评语等部分，因此，成绩单不仅是学生学术能力的证明，也是评估学生是否适合某个教育项目的重要依据！我们承诺采用的是学校原版纸张（原版纸质、底色、纹路）我们工厂拥有全套进口原装设备，特殊工艺都是采用不同机器制作，仿真度基本可以达到100%，所有成品以及工艺效果都可提前给客户展示，不满意可以根据客户要求进行调整，直到满意为止！【主营项目】一、工作未确定，回国需先给父母、亲戚朋友看下文凭的情况，办理毕业证|办理文凭: 买大学毕业证|买大学文凭【q薇1954292140】威斯康星大学河城分校学位证明书如何办理申请？二、回国进私企、外企、自己做生意的情况，这些单位是不查询毕业证真伪的，而且国内没有渠道去查询国外文凭的真假，也不需要提供真实教育部认证。鉴于此，办理美国成绩单威斯康星大学河城分校毕业证【q薇1954292140】国外大学毕业证, 文凭办理, 国外文凭办理, 留信网认证

Crafting-Research-Recommendations Grade 12.pptxDaryllWhere

The Influence off Flexible Work Policiessales480687

This topic explores how flexible work policies—such as remote work, flexible hours, and hybrid models—are transforming modern workplaces. It examines the impact on employee productivity, job satisfaction, work-life balance, and organizational performance. The topic also addresses challenges such as communication gaps, maintaining company culture, and ensuring accountability. Additionally, it highlights how flexible work arrangements can attract top talent, promote inclusivity, and adapt businesses to an evolving global workforce. Ultimately, it reflects the shift in how and where work gets done in the 21st century.

最新版美国芝加哥大学毕业证（鲍颁丑颈肠补驳辞毕业证书）原版定制taqyea

2025原版芝加哥大学毕业证书pdf电子版【q薇1954292140】美国毕业证办理UChicago芝加哥大学毕业证书多少钱？【q薇1954292140】海外各大学Diploma版本，因为疫情学校推迟发放证书、证书原件丢失补办、没有正常毕业未能认证学历面临就业提供解决办法。当遭遇挂科、旷课导致无法修满学分，或者直接被学校退学，最后无法毕业拿不到毕业证。此时的你一定手足无措，因为留学一场，没有获得毕业证以及学历证明肯定是无法给自己和父母一个交代的。【复刻芝加哥大学成绩单信封,Buy The University of Chicago Transcripts】购买日韩成绩单、英国大学成绩单、美国大学成绩单、澳洲大学成绩单、加拿大大学成绩单（q微1954292140）新加坡大学成绩单、新西兰大学成绩单、爱尔兰成绩单、西班牙成绩单、德国成绩单。成绩单的意义主要体现在证明学习能力、评估学术背景、展示综合素质、提高录取率，以及是作为留信认证申请材料的一部分。芝加哥大学成绩单能够体现您的的学习能力，包括芝加哥大学课程成绩、专业能力、研究能力。（q微1954292140）具体来说，成绩报告单通常包含学生的学习技能与习惯、各科成绩以及老师评语等部分，因此，成绩单不仅是学生学术能力的证明，也是评估学生是否适合某个教育项目的重要依据！我们承诺采用的是学校原版纸张（原版纸质、底色、纹路）我们工厂拥有全套进口原装设备，特殊工艺都是采用不同机器制作，仿真度基本可以达到100%，所有成品以及工艺效果都可提前给客户展示，不满意可以根据客户要求进行调整，直到满意为止！【主营项目】一、工作未确定，回国需先给父母、亲戚朋友看下文凭的情况，办理毕业证|办理文凭: 买大学毕业证|买大学文凭【q薇1954292140】芝加哥大学学位证明书如何办理申请？二、回国进私企、外企、自己做生意的情况，这些单位是不查询毕业证真伪的，而且国内没有渠道去查询国外文凭的真假，也不需要提供真实教育部认证。鉴于此，办理美国成绩单芝加哥大学毕业证【q薇1954292140】国外大学毕业证, 文凭办理, 国外文凭办理, 留信网认证

英国毕业证范本利物浦约翰摩尔斯大学成绩单底纹防伪尝闯惭鲍学生证办理学历认证 taqyed

LJMU利物浦约翰摩尔斯大学毕业证书多少钱【q薇1954292140】1:1原版利物浦约翰摩尔斯大学毕业证+LJMU成绩单【q薇1954292140】完美还原海外各大学毕业材料上的工艺：水印，阴影底纹，钢印LOGO烫金烫银，LOGO烫金烫银复合重叠。文字图案浮雕、激光镭射、紫外荧光、温感、复印防伪等防伪工艺。【主营项目】一、工作未确定，回国需先给父母、亲戚朋友看下文凭的情况，办理毕业证|办理文凭: 买大学毕业证|买大学文凭【q薇1954292140】学位证明书如何办理申请？二、回国进私企、外企、自己做生意的情况，这些单位是不查询毕业证真伪的，而且国内没有渠道去查询国外文凭的真假，也不需要提供真实教育部认证。鉴于此，办理利物浦约翰摩尔斯大学毕业证|LJMU成绩单【q薇1954292140】国外大学毕业证, 文凭办理, 国外文凭办理, 留信网认证三.材料咨询办理、认证咨询办理请加学历顾问【微信:1954292140】毕业证购买指大学文凭购买，毕业证办理和文凭办理。学院文凭定制，学校原版文凭补办，扫描件文凭定做，100%文凭复刻。

YEAP !NOT WHAT YOU THINK aakshdjdncnkenfjpayalmistryb

Skdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnneodnrodndocndodnd0dndjxoxnxndkxnxkdndkxndkdjndnrnidnroz doendodnrodnxkdnrocngksjrndkdnr dnkxnddnkxndnrkdnxnxkdSkdnne

Allotted-MBBS-Student-list-batch-2021.pdfsubhansaifi0603

Presentation by Tariq & Mohammed (1).pptxAbooddSandoqaa

美国毕业证范本中华盛顿大学学位证书颁奥鲍学生卡购买Taqyea

UPS and Big Data intro to Business Analytics.pptxsanjum5582

presentation4.pdf Intro to mcmc methodssSergeyTsygankov6

Microsoft Power BI - Advanced Certificate for Business Intelligence using Pow...Prasenjit Debnath

NASA ESE Study Results v4 05.29.2020.pptxCiroAlejandroCamacho

NVIDIA Triton Inference Server, a game-changing platform for deploying AI mod...Tamanna36

25 items quiz for practical research 1 in grade 11leamaydayaganon81

Indigo dyeing Presentation (2).pptx as dyeshreeroop1335

Data Visualisation in data science for studentsconfidenceascend

MRI Pulse Sequence in radiology physics.pptxBelaynehBishaw

Indigo_Airlines_Strategy_Presentation.pptxmukeshpurohit991

Camuflaje Tipos Características Militar 2025.ppte58650738

最新版美国威斯康星大学河城分校毕业证（鲍奥搁贵毕业证书）原版定制taqyea

Crafting-Research-Recommendations Grade 12.pptxDaryllWhere

The Influence off Flexible Work Policiessales480687

最新版美国芝加哥大学毕业证（鲍颁丑颈肠补驳辞毕业证书）原版定制taqyea

英国毕业证范本利物浦约翰摩尔斯大学成绩单底纹防伪尝闯惭鲍学生证办理学历认证 taqyed

YEAP !NOT WHAT YOU THINK aakshdjdncnkenfjpayalmistryb

Allotted-MBBS-Student-list-batch-2021.pdfsubhansaifi0603

Presentation by Tariq & Mohammed (1).pptxAbooddSandoqaa

Data Science Projects @ Runnr

1. Data Science Talk AMOL SAHASRABUDHE ANKIT JAIN

2. WHY THIS PRESENTATION DEMOCRATIZE BRAINSTORMING 2

3. ? Why and What of Data Science ? Data Science @Roadrunnr ? Multi Drop Logic ? Demand Prediction ? Future Work Agenda 3

4. What is Data Science ? 4

5. Data is Everywhere 5

6. HOUSE OF CARDS MAKING SENSE OF DATA 6

7. HOUSE OF CARDS WHO USES DATA SCIENCE 7

8. HOUSE OF CARDS WHO IS A DATA SCIENTIST 8

9. HOUSE OF CARDS WHAT DO THEY DO 9

10. HOUSE OF CARDS WHAT DO THEY DO IDEA EXPERIMENT VALIDATE SPEC DEPLOY MODELING SKILLS : Math, Statistics, Programming and Domain Knowledge 10

11. HOUSE OF CARDS DATA SCIENCE@ROADRUNNR 11

12. Use data as a raw input to develop products to: ? Minimize Estimated Time of Arrival (ETA) ? Improve Reliability at minimal Costs VALUES 12

13. Multi Drop Clubbing Logic ? Improve the multi drop grouping logic to: ? Reduce number of touchpoints/order (5-15%) ? Reduce distance travelled per touch point ? Increase driver satisfaction 13

14. Existing Multi Drop Logic HUB Red Cluster can be done away with in this scenario 14

15. Ideal Clubbing HUB DB should be able to cover points on his way 15

16. How to Achieve Ideal Clubbing? HUB D1 < D2 Join two points which have minimum distance between them ? Joining two points 16

17. Joining Two Groups ? Join on the basis of shortest distances and not center distances to achieve ideal clubbing HUB ? S1, S2 shortest distances ? C1, C2 center distances 17

18. Delhi West Hub Orders (Existing Clubbing) 18

19. Delhi West Hub Orders (New Clubbing) 19

20. Algorithm Limitations ? Sub optimal solution due to accuracy vs complexity trade off ? Performance limited by accuracy of Lat,Long of drop points ? Not optimized for size and weight of shipments ? Routing not included as a part of this version (V2) ? Considers only bike as a carrier (V2) 20

21. Demand Prediction 21

22. Demand Prediction (Impact) ?30% stock-outs during peak hours ?70% demand during peak hours ?Supply Planning ?Driver Placement ?Surge Pricing 22

23. Demand Prediction(Actual Demand) Koramangala High Variance in hourly demand 23

24. Demand Prediction(Method) 24

25. Demand Prediction(Actual and Predicted) Koramangala 25

26. Demand Prediction(drill down) Koramangala 8PM Predictability in number of orders in this hour 26

27. Demand Prediction(Model Failure) Electronic City Demand is too erratic to be captured by modeling Bad predictions 27

28. Demand Prediction(Future Work) ?Category specific demand ?Prediction for smaller clusters rather than localities ?Prediction over smaller intervals ?Stock-out incorporation 28

29. Data Science Future Projects ?Surge pricing ?Wait time prediction for food orders (deployment pending) ?Carrier Selection (Extension of multi drop logic) ?Driver Fraud detection ?Product size estimation using image 29

30. THANK YOU 30

狠狠撸

Data Science Projects @ Runnr

Recommended

More Related Content

Similar to Data Science Projects @ Runnr (20)

Recently uploaded (20)

Data Science Projects @ Runnr