�ݺ�ߣshows by User: lundjohnson

�ݺ�ߣshows by User: lundjohnson / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: lundjohnson / Mon, 05 Jun 2017 03:25:20 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: lundjohnson Graph Walks & Vector Embeddings: Exploiting the head and exploring the tail /slideshow/20170605-mlsfgraph-embedbrianjohnsonv2/76643538 2017-170605032520
Pinterest has the world’s largest catalog of human curated ideas. We’re building a visual discovery engine with 100+ billion ideas, collected by 175+ million people worldwide. As we work to match the right Pin to the right person at the right time, personalization is crucial. Random graph walks with restart are an excellent way to surface popular, high quality, relevant content. But we can also show you great ideas you may not even have known you were looking for - and that’s where vector embedding comes in. We embed you and these billions of ideas in a 128 or 256 dimensional space. Then we project them down into 1000 bits, cut them up into 16 bit chunks, index these chunks, and then find these ideas for you really fast using core search technology. Bio Brian joined Pinterest in 2017 as the Head of Knowledge. He was previously at eBay, Handspring, Excite@Home, Synopsys, and AT&T Bell Labs. Brian received his Ph.D. in Computer Science from the University of Maryland. His original Treemap data visualization paper has been cited thousands of times. ]]>
Pinterest has the world’s largest catalog of human curated ideas. We’re building a visual discovery engine with 100+ billion ideas, collected by 175+ million people worldwide. As we work to match the right Pin to the right person at the right time, personalization is crucial. Random graph walks with restart are an excellent way to surface popular, high quality, relevant content. But we can also show you great ideas you may not even have known you were looking for - and that’s where vector embedding comes in. We embed you and these billions of ideas in a 128 or 256 dimensional space. Then we project them down into 1000 bits, cut them up into 16 bit chunks, index these chunks, and then find these ideas for you really fast using core search technology. Bio Brian joined Pinterest in 2017 as the Head of Knowledge. He was previously at eBay, Handspring, Excite@Home, Synopsys, and AT&T Bell Labs. Brian received his Ph.D. in Computer Science from the University of Maryland. His original Treemap data visualization paper has been cited thousands of times. ]]> Mon, 05 Jun 2017 03:25:20 GMT /slideshow/20170605-mlsfgraph-embedbrianjohnsonv2/76643538 lundjohnson@slideshare.net(lundjohnson) Graph Walks & Vector Embeddings: Exploiting the head and exploring the tail lundjohnson Pinterest has the world’s largest catalog of human curated ideas. We’re building a visual discovery engine with 100+ billion ideas, collected by 175+ million people worldwide. As we work to match the right Pin to the right person at the right time, personalization is crucial. Random graph walks with restart are an excellent way to surface popular, high quality, relevant content. But we can also show you great ideas you may not even have known you were looking for - and that’s where vector embedding comes in. We embed you and these billions of ideas in a 128 or 256 dimensional space. Then we project them down into 1000 bits, cut them up into 16 bit chunks, index these chunks, and then find these ideas for you really fast using core search technology. Bio Brian joined Pinterest in 2017 as the Head of Knowledge. He was previously at eBay, Handspring, Excite@Home, Synopsys, and AT&T Bell Labs. Brian received his Ph.D. in Computer Science from the University of Maryland. His original Treemap data visualization paper has been cited thousands of times. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2017-170605032520-thumbnail.jpg?width=120&height=120&fit=bounds" /> Pinterest has the world’s largest catalog of human curated ideas. We’re building a visual discovery engine with 100+ billion ideas, collected by 175+ million people worldwide. As we work to match the right Pin to the right person at the right time, personalization is crucial. Random graph walks with restart are an excellent way to surface popular, high quality, relevant content. But we can also show you great ideas you may not even have known you were looking for - and that’s where vector embedding comes in. We embed you and these billions of ideas in a 128 or 256 dimensional space. Then we project them down into 1000 bits, cut them up into 16 bit chunks, index these chunks, and then find these ideas for you really fast using core search technology. Bio Brian joined Pinterest in 2017 as the Head of Knowledge. He was previously at eBay, Handspring, Excite@Home, Synopsys, and AT&T Bell Labs. Brian received his Ph.D. in Computer Science from the University of Maryland. His original Treemap data visualization paper has been cited thousands of times.

Graph Walks & Vector Embeddings: Exploiting the head and exploring the tail from Brian Johnson

]]> 1180 2 https://cdn.slidesharecdn.com/ss_thumbnails/2017-170605032520-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 eBay Search Query Intent /slideshow/ebay-query-intent/47539083 2015-04-ebayqueryintent-150428201623-conversion-gate01
eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch.]]>
eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch.]]> Tue, 28 Apr 2015 20:16:23 GMT /slideshow/ebay-query-intent/47539083 lundjohnson@slideshare.net(lundjohnson) eBay Search Query Intent lundjohnson eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebayqueryintent-150428201623-conversion-gate01-thumbnail.jpg?width=120&height=120&fit=bounds" /> eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch.

eBay Search Query Intent from Brian Johnson

]]> 945 5 https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebayqueryintent-150428201623-conversion-gate01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 2015-04 eBay Statistics /slideshow/2015-04-ebaystatistics/47341404 2015-04-ebaystatistics-150423120726-conversion-gate02
A very short overview of eBay and an explanation of why statistics are important. Intended for a high school AP Statistics presentation.]]>
A very short overview of eBay and an explanation of why statistics are important. Intended for a high school AP Statistics presentation.]]> Thu, 23 Apr 2015 12:07:25 GMT /slideshow/2015-04-ebaystatistics/47341404 lundjohnson@slideshare.net(lundjohnson) 2015-04 eBay Statistics lundjohnson A very short overview of eBay and an explanation of why statistics are important. Intended for a high school AP Statistics presentation. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebaystatistics-150423120726-conversion-gate02-thumbnail.jpg?width=120&height=120&fit=bounds" /> A very short overview of eBay and an explanation of why statistics are important. Intended for a high school AP Statistics presentation.

2015-04 eBay Statistics from Brian Johnson

]]> 919 12 https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebaystatistics-150423120726-conversion-gate02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 eBay Search Science, IEEE Big Data, April 3rd, 2015 /slideshow/2015-04e-bayieeebigdata/46599378 2015-04-ebayieeebigdata-150402212922-conversion-gate01
Topic: eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch. Time permitting he will also touch on statistical machine translation based spell correction and machine learned search spam detection. ]]>
Topic: eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch. Time permitting he will also touch on statistical machine translation based spell correction and machine learned search spam detection. ]]> Thu, 02 Apr 2015 21:29:21 GMT /slideshow/2015-04e-bayieeebigdata/46599378 lundjohnson@slideshare.net(lundjohnson) eBay Search Science, IEEE Big Data, April 3rd, 2015 lundjohnson Topic: eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch. Time permitting he will also touch on statistical machine translation based spell correction and machine learned search spam detection. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebayieeebigdata-150402212922-conversion-gate01-thumbnail.jpg?width=120&height=120&fit=bounds" /> Topic: eBay Search Science: Leveraging Behavioral Data Analysis for Effective Query Reformulation Brian will talk about combing through behavioral log files with Scala on Hadoop in order to generate large data sets used to drive dynamic, online query rewrites at eBay. He’ll cover the product/feature pipeline from ideation to data mining, prototyping, statistical analysis, offline side by side analysis, human judgment, online experimentation, and finally launch. Time permitting he will also touch on statistical machine translation based spell correction and machine learned search spam detection.

eBay Search Science, IEEE Big Data, April 3rd, 2015 from Brian Johnson

]]> 3674 9 https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebayieeebigdata-150402212922-conversion-gate01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 CloudCon Data Mining Presentation /slideshow/cloudcon-data-mining-presentation/14561095 2012-10-cloudcon-121002134110-phpapp01
]]>
]]> Tue, 02 Oct 2012 13:41:07 GMT /slideshow/cloudcon-data-mining-presentation/14561095 lundjohnson@slideshare.net(lundjohnson) CloudCon Data Mining Presentation lundjohnson <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2012-10-cloudcon-121002134110-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" />

CloudCon Data Mining Presentation from Brian Johnson

]]> 568 2 https://cdn.slidesharecdn.com/ss_thumbnails/2012-10-cloudcon-121002134110-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 2011 Crowdsourcing Search Evaluation /slideshow/2011-crowdsourcing-search-evaluation-9702217/9702217 2011-09-crowdsourcingsearchevaluation-111014162430-phpapp02
]]>
]]> Fri, 14 Oct 2011 16:24:30 GMT /slideshow/2011-crowdsourcing-search-evaluation-9702217/9702217 lundjohnson@slideshare.net(lundjohnson) 2011 Crowdsourcing Search Evaluation lundjohnson <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2011-09-crowdsourcingsearchevaluation-111014162430-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" />

2011 Crowdsourcing Search Evaluation from Brian Johnson

]]> 495 2 https://cdn.slidesharecdn.com/ss_thumbnails/2011-09-crowdsourcingsearchevaluation-111014162430-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 2011 x.commerce Innovate Data Alchemy /slideshow/2011-xcommerce-innovate-data-alchemy/9701717 2011-10-innovatedataalchemy-111014151651-phpapp01
The New Alchemy: Turning Data into Gold Developers are leading the charge to turn consumer behavior into profitable solutions. By accessing and analyzing the explosion of data from consumer activities, any developer can create the personalized, relevant products and services that customers demand and merchants urgently need. We will discuss how to acquire, store, and mine information, and how to design analytics-focused software and build data-driven software engines. ]]>
The New Alchemy: Turning Data into Gold Developers are leading the charge to turn consumer behavior into profitable solutions. By accessing and analyzing the explosion of data from consumer activities, any developer can create the personalized, relevant products and services that customers demand and merchants urgently need. We will discuss how to acquire, store, and mine information, and how to design analytics-focused software and build data-driven software engines. ]]> Fri, 14 Oct 2011 15:16:49 GMT /slideshow/2011-xcommerce-innovate-data-alchemy/9701717 lundjohnson@slideshare.net(lundjohnson) 2011 x.commerce Innovate Data Alchemy lundjohnson The New Alchemy: Turning Data into Gold Developers are leading the charge to turn consumer behavior into profitable solutions. By accessing and analyzing the explosion of data from consumer activities, any developer can create the personalized, relevant products and services that customers demand and merchants urgently need. We will discuss how to acquire, store, and mine information, and how to design analytics-focused software and build data-driven software engines. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/2011-10-innovatedataalchemy-111014151651-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> The New Alchemy: Turning Data into Gold Developers are leading the charge to turn consumer behavior into profitable solutions. By accessing and analyzing the explosion of data from consumer activities, any developer can create the personalized, relevant products and services that customers demand and merchants urgently need. We will discuss how to acquire, store, and mine information, and how to design analytics-focused software and build data-driven software engines.

2011 x.commerce Innovate Data Alchemy from Brian Johnson

]]> 352 2 https://cdn.slidesharecdn.com/ss_thumbnails/2011-10-innovatedataalchemy-111014151651-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 Treemaps: Visualizing Hierarchical and Categorical Data /slideshow/treemaps-visualizing-hierarchical-and-categorical-data/8721106 1993brianjohnsondissertation-1311918672257-phpapp02-110729005208-phpapp02
1993 Brian Johnson Dissertation]]>
1993 Brian Johnson Dissertation]]> Fri, 29 Jul 2011 00:51:52 GMT /slideshow/treemaps-visualizing-hierarchical-and-categorical-data/8721106 lundjohnson@slideshare.net(lundjohnson) Treemaps: Visualizing Hierarchical and Categorical Data lundjohnson 1993 Brian Johnson Dissertation <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/1993brianjohnsondissertation-1311918672257-phpapp02-110729005208-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> 1993 Brian Johnson Dissertation

Treemaps: Visualizing Hierarchical and Categorical Data from Brian Johnson

]]> 3230 6 https://cdn.slidesharecdn.com/ss_thumbnails/1993brianjohnsondissertation-1311918672257-phpapp02-110729005208-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 11 964 181 System And Method For Providi /slideshow/11-964-181-system-and-method-for-providi/8720918 11964181systemandmethodforprovidi-13119174074479-phpapp01-110729004934-phpapp01
Seller Tagging Patent]]>
Seller Tagging Patent]]> Fri, 29 Jul 2011 00:33:21 GMT /slideshow/11-964-181-system-and-method-for-providi/8720918 lundjohnson@slideshare.net(lundjohnson) 11 964 181 System And Method For Providi lundjohnson Seller Tagging Patent <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/11964181systemandmethodforprovidi-13119174074479-phpapp01-110729004934-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> Seller Tagging Patent

11 964 181 System And Method For Providi from Brian Johnson

]]> 287 1 https://cdn.slidesharecdn.com/ss_thumbnails/11964181systemandmethodforprovidi-13119174074479-phpapp01-110729004934-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 11 641 262 Proprietor Currency Assignmen /slideshow/11-641-262-proprietor-currency-assignmen/8720917 11641262proprietorcurrencyassignmen-131191739675-phpapp02-110729005000-phpapp02
Media Swapping Patent]]>
Media Swapping Patent]]> Fri, 29 Jul 2011 00:33:19 GMT /slideshow/11-641-262-proprietor-currency-assignmen/8720917 lundjohnson@slideshare.net(lundjohnson) 11 641 262 Proprietor Currency Assignmen lundjohnson Media Swapping Patent <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/11641262proprietorcurrencyassignmen-131191739675-phpapp02-110729005000-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> Media Swapping Patent

11 641 262 Proprietor Currency Assignmen from Brian Johnson

]]> 308 1 https://cdn.slidesharecdn.com/ss_thumbnails/11641262proprietorcurrencyassignmen-131191739675-phpapp02-110729005000-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 10 977 279 Method And System For Categor /slideshow/10-977-279-method-and-system-for-categor/8720915 10977279methodandsystemforcategor-13119173931297-phpapp01-110729003405-phpapp01
Category Redirection Patent]]>
Category Redirection Patent]]> Fri, 29 Jul 2011 00:33:17 GMT /slideshow/10-977-279-method-and-system-for-categor/8720915 lundjohnson@slideshare.net(lundjohnson) 10 977 279 Method And System For Categor lundjohnson Category Redirection Patent <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/10977279methodandsystemforcategor-13119173931297-phpapp01-110729003405-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> Category Redirection Patent

10 977 279 Method And System For Categor from Brian Johnson

]]> 361 1 https://cdn.slidesharecdn.com/ss_thumbnails/10977279methodandsystemforcategor-13119173931297-phpapp01-110729003405-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 11 869 290 Electronic Publication System /slideshow/11-869-290-electronic-publication-system/8720914 11869290electronicpublicationsystem-13119174011492-phpapp02-110729005203-phpapp02
Seller Tagging Patent]]>
Seller Tagging Patent]]> Fri, 29 Jul 2011 00:33:14 GMT /slideshow/11-869-290-electronic-publication-system/8720914 lundjohnson@slideshare.net(lundjohnson) 11 869 290 Electronic Publication System lundjohnson Seller Tagging Patent <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/11869290electronicpublicationsystem-13119174011492-phpapp02-110729005203-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> Seller Tagging Patent

11 869 290 Electronic Publication System from Brian Johnson

]]> 299 1 https://cdn.slidesharecdn.com/ss_thumbnails/11869290electronicpublicationsystem-13119174011492-phpapp02-110729005203-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds document Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 2011 Search Query Rewrites - Synonyms & Acronyms /slideshow/2011-07-27-bay-area-search/8720913 20110727bayareasearch-13119174449595-phpapp01-110729003404-phpapp01
July 27, 2011 Bay Area Search Presentation Brian Johnson, Engineering Director, Query Services @ eBay Query expansion is an important part of of the search recall for all search engines. In this talk I'll discuss some of the general trend driving Hadoop adoption within the Search Query Services team at eBay, and the types of algorithms/techniques we've moved to Hadoop at eBay. Over time we've moved from smaller, editorial data sets to large machine generated data sets mined from behavior log data, items/listings, catalogs, etc. One common workflow is to mine large candidate rewrites/expansions data sets from multiple data sources, use crowd sourced human judgment to classify a subset of the candidates (true positive, false positive), use machine learning techniques discard false positives, run automated validation on the final data set, and automatically push to production. Ravi Jammalakadaka, Senior Applied Researcher, Query Services @ eBay Ravi is a real engineer. Not a pointy haired manager like the previous speaker. Expect some real engineering:-) He'll be doing a literature review for acronym mining and discussing a real world implementation. Title: Mining Acronyms From Raw Text Abstract: Significant number of eBay products are known by their acronyms. eBay query expansion service expands user queries by their acronym equivalents to increase recall. The challenge is to mine acronyms from either seller ( ex. item descriptions, titles) or buyer ( ex. queries) data. Ravi will present the state of the art algorithms from recent conferences that mine acronyms from raw text and present their limitations. He will present a new acronym mining algorithm that seeks to address the limitations identified with previous algorithms. He will present a machine learning classifier that seeks to remove the false positives generated from the acronym mining algorithm. ]]>
July 27, 2011 Bay Area Search Presentation Brian Johnson, Engineering Director, Query Services @ eBay Query expansion is an important part of of the search recall for all search engines. In this talk I'll discuss some of the general trend driving Hadoop adoption within the Search Query Services team at eBay, and the types of algorithms/techniques we've moved to Hadoop at eBay. Over time we've moved from smaller, editorial data sets to large machine generated data sets mined from behavior log data, items/listings, catalogs, etc. One common workflow is to mine large candidate rewrites/expansions data sets from multiple data sources, use crowd sourced human judgment to classify a subset of the candidates (true positive, false positive), use machine learning techniques discard false positives, run automated validation on the final data set, and automatically push to production. Ravi Jammalakadaka, Senior Applied Researcher, Query Services @ eBay Ravi is a real engineer. Not a pointy haired manager like the previous speaker. Expect some real engineering:-) He'll be doing a literature review for acronym mining and discussing a real world implementation. Title: Mining Acronyms From Raw Text Abstract: Significant number of eBay products are known by their acronyms. eBay query expansion service expands user queries by their acronym equivalents to increase recall. The challenge is to mine acronyms from either seller ( ex. item descriptions, titles) or buyer ( ex. queries) data. Ravi will present the state of the art algorithms from recent conferences that mine acronyms from raw text and present their limitations. He will present a new acronym mining algorithm that seeks to address the limitations identified with previous algorithms. He will present a machine learning classifier that seeks to remove the false positives generated from the acronym mining algorithm. ]]> Fri, 29 Jul 2011 00:33:11 GMT /slideshow/2011-07-27-bay-area-search/8720913 lundjohnson@slideshare.net(lundjohnson) 2011 Search Query Rewrites - Synonyms & Acronyms lundjohnson July 27, 2011 Bay Area Search Presentation Brian Johnson, Engineering Director, Query Services @ eBay Query expansion is an important part of of the search recall for all search engines. In this talk I'll discuss some of the general trend driving Hadoop adoption within the Search Query Services team at eBay, and the types of algorithms/techniques we've moved to Hadoop at eBay. Over time we've moved from smaller, editorial data sets to large machine generated data sets mined from behavior log data, items/listings, catalogs, etc. One common workflow is to mine large candidate rewrites/expansions data sets from multiple data sources, use crowd sourced human judgment to classify a subset of the candidates (true positive, false positive), use machine learning techniques discard false positives, run automated validation on the final data set, and automatically push to production. Ravi Jammalakadaka, Senior Applied Researcher, Query Services @ eBay Ravi is a real engineer. Not a pointy haired manager like the previous speaker. Expect some real engineering:-) He'll be doing a literature review for acronym mining and discussing a real world implementation. Title: Mining Acronyms From Raw Text Abstract: Significant number of eBay products are known by their acronyms. eBay query expansion service expands user queries by their acronym equivalents to increase recall. The challenge is to mine acronyms from either seller ( ex. item descriptions, titles) or buyer ( ex. queries) data. Ravi will present the state of the art algorithms from recent conferences that mine acronyms from raw text and present their limitations. He will present a new acronym mining algorithm that seeks to address the limitations identified with previous algorithms. He will present a machine learning classifier that seeks to remove the false positives generated from the acronym mining algorithm. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/20110727bayareasearch-13119174449595-phpapp01-110729003404-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> July 27, 2011 Bay Area Search Presentation Brian Johnson, Engineering Director, Query Services @ eBay Query expansion is an important part of of the search recall for all search engines. In this talk I'll discuss some of the general trend driving Hadoop adoption within the Search Query Services team at eBay, and the types of algorithms/techniques we've moved to Hadoop at eBay. Over time we've moved from smaller, editorial data sets to large machine generated data sets mined from behavior log data, items/listings, catalogs, etc. One common workflow is to mine large candidate rewrites/expansions data sets from multiple data sources, use crowd sourced human judgment to classify a subset of the candidates (true positive, false positive), use machine learning techniques discard false positives, run automated validation on the final data set, and automatically push to production. Ravi Jammalakadaka, Senior Applied Researcher, Query Services @ eBay Ravi is a real engineer. Not a pointy haired manager like the previous speaker. Expect some real engineering:-) He'll be doing a literature review for acronym mining and discussing a real world implementation. Title: Mining Acronyms From Raw Text Abstract: Significant number of eBay products are known by their acronyms. eBay query expansion service expands user queries by their acronym equivalents to increase recall. The challenge is to mine acronyms from either seller ( ex. item descriptions, titles) or buyer ( ex. queries) data. Ravi will present the state of the art algorithms from recent conferences that mine acronyms from raw text and present their limitations. He will present a new acronym mining algorithm that seeks to address the limitations identified with previous algorithms. He will present a machine learning classifier that seeks to remove the false positives generated from the acronym mining algorithm.

2011 Search Query Rewrites - Synonyms & Acronyms from Brian Johnson

]]> 4699 16 https://cdn.slidesharecdn.com/ss_thumbnails/20110727bayareasearch-13119174449595-phpapp01-110729003404-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

0 https://cdn.slidesharecdn.com/profile-photo-lundjohnson-48x48.jpg?cb=1656374523 Directly improving eBay revenue by millions of dollars per engineer per year via improvements to eBay search recall without degrading relevance. Using data mining, machine learning, and map/reduce hadoop. * Spell Correction * Query Rewrites/Reformulation ** Synonym expansions ** Acronym expansions ** Compound word and space synonym expansion ** Attribute expansions ** Category expansions * Null/Low Recovery * Related Search (2012) Removing Search SPAM (Search Manipulation) * Created new team to combat search engine manipulation (spam) * 2012 Critical Talent & Engineering Champion awards * Responsible for search recall, query rewrites/expansions, spell correction, spam, and related se... ebay.com https://cdn.slidesharecdn.com/ss_thumbnails/2017-170605032520-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/20170605-mlsfgraph-embedbrianjohnsonv2/76643538 Graph Walks & Vector E... https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebayqueryintent-150428201623-conversion-gate01-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/ebay-query-intent/47539083 eBay Search Query Intent https://cdn.slidesharecdn.com/ss_thumbnails/2015-04-ebaystatistics-150423120726-conversion-gate02-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/2015-04-ebaystatistics/47341404 2015-04 eBay Statistics