Upgrade to Pro
— share decks privately, control downloads, hide ads and more …
Speaker Deck
Features
Speaker Deck
PRO
Sign in
Sign up for free
Search
Search
06 Models for Use-Cases
Search
LiLa'16
March 20, 2016
Research
0
54
06 Models for Use-Cases
LiLa'16
March 20, 2016
Tweet
Share
More Decks by LiLa'16
See All by LiLa'16
01 Introduction
lila16ecir
0
73
02 Online Evaluation
lila16ecir
0
70
03 LL4IR Architecture
lila16ecir
0
62
04 Use-Cases
lila16ecir
0
67
05 API Interactions I.
lila16ecir
0
60
07 API Interactions II.
lila16ecir
0
52
08 Interpreting Feedback
lila16ecir
0
85
09 API Interactions III.
lila16ecir
0
49
10 Simulations
lila16ecir
0
71
Other Decks in Research
See All in Research
秘伝:脆弱性診断をうまく活用してセキュリティを確保するには
okdt
PRO
3
740
クラウドソーシングによる学習データ作成と品質管理(セキュリティキャンプ2024全国大会D2講義資料)
takumi1001
0
280
Weekly AI Agents News! 8月号 論文のアーカイブ
masatoto
1
180
医療支援AI開発における臨床と情報学の連携を円滑に進めるために
moda0
0
110
外積やロドリゲスの回転公式を利用した点群の回転
kentaitakura
1
650
Tietovuoto Social Design Agency (SDA) -trollitehtaasta
hponka
0
2.5k
marukotenant01/tenant-20240916
marketing2024
0
500
snlp2024_multiheadMoE
takase
0
430
20240918 交通くまもとーく 未来の鉄道網編(太田恒平)
trafficbrain
0
220
Weekly AI Agents News! 9月号 論文のアーカイブ
masatoto
1
120
Global Evidence Summit (GES) 参加報告
daimoriwaki
0
150
Leveraging LLMs for Unsupervised Dense Retriever Ranking (SIGIR 2024)
kampersanda
2
190
Featured
See All Featured
個人開発の失敗を避けるイケてる考え方 / tips for indie hackers
panda_program
93
16k
A Modern Web Designer's Workflow
chriscoyier
693
190k
How to train your dragon (web standard)
notwaldorf
88
5.7k
Let's Do A Bunch of Simple Stuff to Make Websites Faster
chriscoyier
506
140k
How to Create Impact in a Changing Tech Landscape [PerfNow 2023]
tammyeverts
47
2.1k
Designing on Purpose - Digital PM Summit 2013
jponch
115
7k
Optimizing for Happiness
mojombo
376
70k
A designer walks into a library…
pauljervisheath
203
24k
Visualizing Your Data: Incorporating Mongo into Loggly Infrastructure
mongodb
42
9.2k
Adopting Sorbet at Scale
ufuk
73
9.1k
Become a Pro
speakerdeck
PRO
25
5k
How GitHub (no longer) Works
holman
310
140k
Transcript
Anne Schuth (Blendle / University of Amsterdam, The Netherlands) Krisztian
Balog (University of Stavanger, Norway) Tutorial at ECIR 2016 in Padua, Italy Models for Use-Cases
Overview • Two use-cases • Product search • Academic literature
search • Simple shared approach • Fielded document representation
Baseline model • Single-field document retrieval • E.g., BM25 score
( d, q ) = X t2q ft,d · (1 + k1) ft,d + k1(1 b + b |d| avgdl ) · idft
Single-field document represenation • Using only product name field •
Using catchall field docid R-d3504 product_name LEGO DUPLO Kreatív láda 10817 docid R-d3504 catchall LEGO DUPLO Kreatív láda 10817 Építőjáték, LEGO LEGO Éld ki kreativitásod minden évszakban LEGO DUPLO elemekkel!
Fielded extension • Multi-field document retrieval, e.g., BM25F score (
d, q ) = X t2q ˜ ft,d k1 + ˜ ft,d · idft ˜ ft,d = X i wi ft,di Bi Combining term frequencies across fields Field weight Soft normalization for field i
Multi-field document representation docid R-d3504 product_name LEGO DUPLO Kreatív láda
10817 brand LEGO DUPLO short_description Éld ki kreativitásod minden évszakban LEGO DUPLO elemekkel! catchall LEGO DUPLO Kreatív láda 10817 Építőjáték, LEGO LEGO Éld ki kreativitásod minden évszakban LEGO DUPLO elemekkel!
Incorporating historical feedback • Combining historical score with retrieval score
• A simple approach: • or