about us
play

About us MICES 2017 / sebastian.russ@tudock.de / About us - PowerPoint PPT Presentation

Sep 14, 2022 •295 likes •749 views

I E M C S About us MICES 2017 / sebastian.russ@tudock.de / About us Development Consulting more... more... more... more... Search Magento Experience People Services michael.wolf@tudock.de Initial analysis FACT-Finder Solr

  1. I E M C S

  2. About us MICES 2017 / sebastian.russ@tudock.de / About us

  3. Development Consulting more... more... more... more... Search Magento Experience People Services michael.wolf@tudock.de Initial analysis FACT-Finder Solr Search management Epoq Fredhopper Frontend analysis Celebros SEMKNOX abdullah.zahir@tudock.de KPI monitoring more... Optimize ranking Evaluate search tools sebastian.russ@tudock.de and many more... MICES 2017 / sebastian.russ@tudock.de / About us

  4. Search management workfmows MICES 2017 / sebastian.russ@tudock.de / Search management workfmows

  5. Search result quality Continuous optimization Initial analysis Status quo Time MICES 2017 / sebastian.russ@tudock.de / Search management workfmows

  6. Initial analysis MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  7. Search result quality Initial analysis Time MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  8. Product data User behaviour Onsite search Search quality MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  9. Data analysis Search term analysis Search configuration Comparison Evaluation / to-dos MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  10. Data analysis MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  11. ID Name Category Size Check: Quality 10001 Jacket Coats & Jackets XL 10002 Jacket Coats & Jackets L Consistency 10003 Leather Jacket Jackets 42 10004 Parka Coats & Jackets M-L Gaps 10005 Parka - M-L 10006 Jacket Jackets S 10007 Coat Jackets M 10008 Jacket Jackets - 10009 Vest Jackets L 10010 Leather Jacket - 44 10011 Jacket Jackets XL 10012 Jacket Jackets XXL 10013 Coat - L 10014 Jacket Jackets L MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis / Data analysis

  12. Search term analysis MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  13. Search term: Clusters: Jacket Category Leather Jacket Attribute Category men‘s jacket xl Attribute Category Attribute winter parka Attribute Category wool coat Attribute Category jacket size 42 Category Attribute shipping Service blue jacket Attribute Category 10009 Vest ID Category grey women coats Attribute Attribute Category brown leather jacket Attribute Attribute Category levis jacket Brand Category 10014 ID Jacket Category MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis / Search term analysis

  14. Search confjguration MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  15. Search configuration Synonyms Ranking Common topics Campaigns Suggest Technology based topics SEMKNOX Celebros Solr Fredhopper Epoq FACT-Finder Ontology Ontology Analyzers Analyzers Bandwith Preprocessors Concepts Concepts Filters Filters Similarity Bandwith Mappings Mappings Query/index time Search passes Similarity Analyzers Analyzers Multi-term synonyms Search passes MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis / Search confjguration

  16. Comparison MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis

  17. Data analysis Search term analysis Search configuration Comparison Evaluation / to-dos MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Initial analysis / Comparison

  18. Continuous optimization MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization

  19. Search result quality Continuous optimization Time MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization

  20. Test and deploy Pick search terms Continuous optimization Optimize search results Check search results MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization

  21. Pick search terms MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization

  22. number of searches diversity search term complexity MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  23. Top searches Longtail MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  24. Top searches Zero Results MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  25. MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  26. MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  27. MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  28. Top searches KPIs Samples Clusters Zero results MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  29. MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  30. MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Pick search terms

  31. Identify trends MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization

  32. Optimization Transactions Click through rate Number of searches time in weeks MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Identify trends

  33. Wasted opportunity Transactions Click through rate Number of searches three weeks later time in weeks MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Identify trends

  34. this week 100% Swimwear Underwear percentage of clustered topics Winter sports 0% Week 01 Week 52 time in weeks MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Identify trends

  35. Clustering search problems MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization

  36. What? Why? Things to document: Problem cluster: Searching for There‘s no information like Date Search term Data „crew neck shirt“ „crew neck“ or similar in Metrics KPIs returns zero results. the product feed. Problem Solution Deployment (date) Searching for There‘s no filter Date Search term Search configuration „levi‘s jeans“ doesn‘t configured to replace Metrics KPIs return the same results the apostrophe in Solr‘s as searching for analyzer chain. Problem Solution „levis jeans“. Deployment (date) Searching for The category is called Date Search term Synonyms „nike apparel“ „clothing“. The synonym Metrics KPIs returns zero results. „clothing“ = „apparel“ is not configured yet. Problem Solution Deployment (date) MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Clustering search problems

  37. Data Synonyms Search configuration MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Clustering search problems

  38. Data Synonyms Search configuration MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Clustering search problems

  39. Data Synonyms Search configuration Ranking Analyzer Suggest MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Clustering search problems

  40. Project A Project B Project C Data Synonyms Search configuration Low data quality Good data quality Average data quality Only basic ranking Poor ranking Well defined ranking Low precision Low precision High precision MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Clustering search problems

  41. Why it‘s worth the effort MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization

  42. Search result quality Optimization freeze - New products - New trends - New technology - Changes in user behavior - Data errors Time Analysis Implementation Continuous optimization MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Continuous optimization / Why it‘s worth the effort

  43. Questions? MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Questions?

  44. Thanks :-) MICES 2017 / sebastian.russ@tudock.de / Search management workfmows / Thanks :-)

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Can Industrial Cluster Strategy Improve the Competitiveness of Industry: Evidence from Nigeria?

Can Industrial Cluster Strategy Improve the Competitiveness of Industry: Evidence from Nigeria?

Can Industrial Cluster Strategy Improve the Competitiveness of Industry: Evidence from Nigeria? Louis N. Chete*, Foluso Adeyinka*, Olorunfemi Ogundele* and Emma Howard** Introduction Cluster development has emerged as an important new

471 views • 34 slides
Signaling Requirements at the Optical UNI draft-bala-mpls-optical-uni-signaling-00.txt O.

Signaling Requirements at the Optical UNI draft-bala-mpls-optical-uni-signaling-00.txt O.

Signaling Requirements at the Optical UNI draft-bala-mpls-optical-uni-signaling-00.txt O. Abul-Magd (Nortel), O. Aparacio (C&W), R. Barry (Sycamore), G. Bernstien (Ciena), R. Jain (Nayna), L. Jia, R. Dulepet (ONI), M. Lazer, J. Yates

813 views • 16 slides
Estimating Urban CO 2 Fluxes at High Spatial Resolution from In Situ Observations Alexis

Estimating Urban CO 2 Fluxes at High Spatial Resolution from In Situ Observations Alexis

Estimating Urban CO 2 Fluxes at High Spatial Resolution from In Situ Observations Alexis Shusterman, UC Berkeley shusterman.alexis@berkeley.edu http://beacon.berkeley.edu BErkeley 465m Atmospheric CO 2 Observation Network 2km ~25

520 views • 15 slides
Why Mixed Effects Models? Mixed Effects Models Recap/Intro Three issues with ANOVA

Why Mixed Effects Models? Mixed Effects Models Recap/Intro Three issues with ANOVA

Why Mixed Effects Models? Mixed Effects Models Recap/Intro Three issues with ANOVA Multiple random effects Categorical data Focus on fixed effects What mixed effects models do Random slopes Link functions Iterative

575 views • 38 slides
Using the New Products Margin to Predict the Industry-Level Impact of Trade Reform Timothy J.

Using the New Products Margin to Predict the Industry-Level Impact of Trade Reform Timothy J.

Using the New Products Margin to Predict the Industry-Level Impact of Trade Reform Timothy J. Kehoe University of Minnesota and Federal Reserve Bank of Minneapolis Jack M. Rossbach University of Minnesota and Federal Reserve Bank of

861 views • 33 slides
Deep Learning of Optimization Heuristics http://chriscummins.cc/pact17 Chris Cummins

Deep Learning of Optimization Heuristics http://chriscummins.cc/pact17 Chris Cummins

End-to-end Deep Learning of Optimization Heuristics http://chriscummins.cc/pact17 Chris Cummins University of Edinburgh Pavlos Petoumenos University of Edinburgh Zheng Wang Lancaster University Hugh Leather University of Edinburgh

447 views • 33 slides
The Golden Rule of Forecasting Kesten C. Green University of South Australia Business School

The Golden Rule of Forecasting Kesten C. Green University of South Australia Business School

The Golden Rule of Forecasting Kesten C. Green University of South Australia Business School with J. Scott Armstrong Wharton, University of Pennsylvania Andreas Graefe LMU Munich, Germany International Symposium on Forecasting, Seoul 11:40

327 views • 28 slides
Conceptual spaces for matching and representing preferences Anton Benz Alexandra Strekalova ZAS

Conceptual spaces for matching and representing preferences Anton Benz Alexandra Strekalova ZAS

Conceptual spaces for matching and representing preferences Anton Benz Alexandra Strekalova ZAS Berlin Tandem Workshop on Optimality in Language and Geometric Approaches to Cognition 13 December 2010 Anton Benz Alexandra Strekalova ZAS

281 views • 25 slides
Data Processing on Modern Hardware Jens Teubner, TU Dortmund, DBIS Group

Data Processing on Modern Hardware Jens Teubner, TU Dortmund, DBIS Group

Data Processing on Modern Hardware Jens Teubner, TU Dortmund, DBIS Group jens.teubner@cs.tu-dortmund.de Summer 2014 Jens Teubner Data Processing on Modern Hardware Summer 2014 c 1 Part VII FPGAs for Data Processing Jens Teubner

971 views • 58 slides
HPC Environment Management: New Challenges in the Petaflop Era Jonas Dias jonas@nacad.ufrj.br

HPC Environment Management: New Challenges in the Petaflop Era Jonas Dias jonas@nacad.ufrj.br

HPC Environment Management: New Challenges in the Petaflop Era Jonas Dias jonas@nacad.ufrj.br Albino Aveleda bino@nacad.ufrj.br VECPAR10 Agenda 1. Introduction 2. Available Tools 1. Deployment 2. Monitoring 3. Proprietary Solutions

584 views • 27 slides
I am NOT: Wishful Thinking Against New Ideas Against New Therapies Lemming

I am NOT: Wishful Thinking Against New Ideas Against New Therapies Lemming

4/14/2016 Disclosures Inferior Vena Cava Filters: A Love /Hate (Mostly Hate) Relationship Lack of Political Correctness Gregory L. Moneta, M.D. Professor and Chief, Vascular Surgery Oregon Health & Science University Knight

369 views • 10 slides
Pumping Lemma for CFLs In any sufficiently long string in a CFL, it is possible to find at most

Pumping Lemma for CFLs In any sufficiently long string in a CFL, it is possible to find at most

Pumping Lemma for CFLs In any sufficiently long string in a CFL, it is possible to find at most two short, nearby sub- strings that we can pump i times in tandem, for any integer i , and the resulting string will still be in that language.

560 views • 4 slides
Dynamic hedging for the real option management of electricity storage Joakim Dimoski, Stein-Erik

Dynamic hedging for the real option management of electricity storage Joakim Dimoski, Stein-Erik

Dynamic hedging for the real option management of electricity storage Joakim Dimoski, Stein-Erik Fleten, Nils Lhndorf and Sveinung Nersten International storage-based energy trading 2 ECN: A clean economy in 2050 requires cross-border

399 views • 24 slides
trt rr rs t

trt rr rs t

trt rr rs t rrt rt rt t ttt t tr

360 views • 21 slides
Was tut ein guter Software Architekt? Eberhard Wolff Architecture and Technology Manager adesso

Was tut ein guter Software Architekt? Eberhard Wolff Architecture and Technology Manager adesso

Was tut ein guter Software Architekt? Eberhard Wolff Architecture and Technology Manager adesso AG 28.03.12 About Me Eberhard Wolff Architecture & Technology Manager at adesso adesso is a leading IT consultancy in the German

667 views • 40 slides
Algorithms Algorithms Gerth Stlting Brodal Overview Overview Algo Algorit rithm hms s gr

Algorithms Algorithms Gerth Stlting Brodal Overview Overview Algo Algorit rithm hms s gr

Master Thesis Preparation Algorithms Algorithms Gerth Stlting Brodal Overview Overview Algo Algorit rithm hms s gr grou oup p at at DAIMI DAIMI Who? Where? Courses Research Mas Maste ter r th thes esis is in

413 views • 13 slides
Lattimer subsets of Art IE ) field function for Fixed & sub extensions of E subsets et Aut

Lattimer subsets of Art IE ) field function for Fixed & sub extensions of E subsets et Aut

of Characters Independence fields Fixed of I : subgroups Lattimer subsets of Art IE ) field function for Fixed & sub extensions of E subsets et Aut ( E ) lie , inclusion reversing ) " contravariant " not too small ) (

436 views • 12 slides
Snow Leopard Permutations, Even Knots, Odd Knots, Janus Knots, and Restricted Catalan Paths Ben

Snow Leopard Permutations, Even Knots, Odd Knots, Janus Knots, and Restricted Catalan Paths Ben

Snow Leopard Permutations, Even Knots, Odd Knots, Janus Knots, and Restricted Catalan Paths Ben Caffrey, Eric Egge*, Greg Michel, Kailee Rubin, Jon Ver Steegh Carleton College May 21, 2015 My Students Ben Greg Jon Kailee A Problem in

818 views • 61 slides
Foundations of Computer Science Last Time Lecture 8 Proofs with Recursive Objects Structural

Foundations of Computer Science Last Time Lecture 8 Proofs with Recursive Objects Structural

Foundations of Computer Science Last Time Lecture 8 Proofs with Recursive Objects Structural Induction: Induction on Recursively Defined Objects Proving an object is not in a recursive set Examples: sets, sequences, trees 1 Recursion. 2

627 views • 4 slides
Visualiza(on Parallel Session Summary Enhanced Open Inventor Viewer for Interactive

Visualiza(on Parallel Session Summary Enhanced Open Inventor Viewer for Interactive

Visualiza(on Parallel Session Summary Enhanced Open Inventor Viewer for Interactive Scene Navigation in Geant4 Viewpoints and scene element navigation Fly-through animation along reference path Frederick Jones, TRIUMF Rastislav

279 views • 7 slides
Multi-Level Active Prediction of Useful Image Annotations Sudheendra Vijayanarasimhan and Kristen

Multi-Level Active Prediction of Useful Image Annotations Sudheendra Vijayanarasimhan and Kristen

Multi-Level Active Prediction of Useful Image Annotations Sudheendra Vijayanarasimhan and Kristen Grauman Department of Computer Sciences University of Texas at Austin Austin, Texas 78712 (svnaras,grauman)@cs.utexas.edu Introduction Visual

873 views • 60 slides
Lecture 1: Introduction to Digital Logic Design CSE 140: Components and Design Techniques for

Lecture 1: Introduction to Digital Logic Design CSE 140: Components and Design Techniques for

Lecture 1: Introduction to Digital Logic Design CSE 140: Components and Design Techniques for Digital Systems Spring 2017 CK Cheng Dept. of Computer Science and Engineering University of California, San Diego 1 Outlines Staff

725 views • 38 slides
2014 Coldwell Banker San Francisco Bay Area Top One Percent #48 Abby Tanem Greenbrae #47 Bill

2014 Coldwell Banker San Francisco Bay Area Top One Percent #48 Abby Tanem Greenbrae #47 Bill

2014 Coldwell Banker San Francisco Bay Area Top One Percent #48 Abby Tanem Greenbrae #47 Bill Johnson Group Greenbrae #46 Don & Trudi Copland Danville #45 Ann Harris Sebastopol #44 Andrea Gordon Oakland Piedmont #43 Everett

696 views • 51 slides
Auror ora a Borealis ealis (Automatic) Music - Adagietto, Symphony n 5 in C minor, Gustav

Auror ora a Borealis ealis (Automatic) Music - Adagietto, Symphony n 5 in C minor, Gustav

Auror ora a Borealis ealis (Automatic) Music - Adagietto, Symphony n 5 in C minor, Gustav Mahler Rideout eout Bay, , Ontar ario io Mount unt Everest rest Dusk Mount untain ain Hut, , Everes rest Lario io Lake Serra ra da Leba,

713 views • 60 slides

More recommend