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
07 API Interactions II.
Search
Sponsored
·
Your Podcast. Everywhere. Effortlessly.
Share. Educate. Inspire. Entertain. You do you. We'll handle the rest.
→
LiLa'16
March 20, 2016
Research
60
0
Share
Embed
Copy iframe code
Copy JS code
Copy link
Start on current slide
07 API Interactions II.
LiLa'16
March 20, 2016
More Decks by LiLa'16
See All by LiLa'16
01 Introduction
lila16ecir
0
87
02 Online Evaluation
lila16ecir
0
93
03 LL4IR Architecture
lila16ecir
0
73
04 Use-Cases
lila16ecir
0
79
05 API Interactions I.
lila16ecir
0
69
06 Models for Use-Cases
lila16ecir
0
64
08 Interpreting Feedback
lila16ecir
0
110
09 API Interactions III.
lila16ecir
0
60
10 Simulations
lila16ecir
0
85
Other Decks in Research
See All in Research
計算情報学研究室(数理情報学第7研究室)2026
tomohirokoana
0
700
言語モデルから言語について語る際に押さえておきたいこと
eumesy
PRO
6
2.5k
COFFEE-Japan PROJECT Impact Report(Uminomukou Coffee)
ontheslope
0
300
論文紹介 "ReSim: Reliable World Simulation for Autonomous Driving"
kogo
0
720
Google Cloud Next 2026 DM Recap Agentic Data Cloudを添えて / Google Cloud Next 2026 DM Recap
nnaka2992
0
100
ros2-perf-multihost: 分散システムにおける客観的なアーキテクチャ評価フレームワーク
takasehideki
0
110
【ローカルAI LT大会】SSE: Stable Static Embedding ー速度低下を伴わず 静的埋め込みモデルの潜在能力を引き出す Dynamic Tanh手法の提案
rikkabotan7
0
100
【Zozo Research 技術共有会】三次元領域の現在と展望
mickey_0226
3
510
「AIとWhyを深堀る」をAIと深堀る
iflection
0
540
Dual Quadric表現を用いた動的物体追跡とRGB-D・IMU制約の密結合によるオドメトリ推定
nanoshimarobot
0
470
JICA QUEST 共創×革新プログラム Impact Report(海ノ向こうコーヒー)
ontheslope
0
260
Apache Gravitinoで実現する Icebergカタログ統合とアクセスの一元化
matsumooon
0
380
Featured
See All Featured
Introduction to Domain-Driven Design and Collaborative software design
baasie
1
920
From Legacy to Launchpad: Building Startup-Ready Communities
dugsong
0
280
The Limits of Empathy - UXLibs8
cassininazir
1
570
New Earth Scene 8
popppiees
3
2.4k
Money Talks: Using Revenue to Get Sh*t Done
nikkihalliwell
0
440
My Coaching Mixtape
mlcsv
0
180
The Director’s Chair: Orchestrating AI for Truly Effective Learning
tmiket
1
230
Chasing Engaging Ingredients in Design
codingconduct
0
250
Understanding Cognitive Biases in Performance Measurement
bluesmoon
32
3k
How To Speak Unicorn (iThemes Webinar)
marktimemedia
1
520
Navigating Weather and Climate Data
rabernat
0
430
Context Engineering - Making Every Token Count
addyosmani
9
1k
Transcript
Anne Schuth (Blendle / University of Amsterdam, The Netherlands) Krisztian
Balog (University of Stavanger, Norway) Tutorial at ECIR 2016 in Padua, Italy API Interactions II.
Outline for API interactions • Obtaining an API key and
signing up for a site • Getting queries and candidate items • Generating and uploading rankings • Obtaining feedback and outcome
Uploading ranking Living Labs API Researcher ranking click Queries Documents
Run endpoint • PUT /api/participant/run/(key)/(qid) • Submit a ranking for
a specific query • A run is just a ranking of docids as shown to users • A run can be updated • Note: runs for test queries can only be updated before the test period starts
Example { 'doclist': [ {'docid': u'R-d70'}, {'docid': u'R-d72'}, {'docid': u'R-d74'},
{'docid': u'R-d75'}, {'docid': u'R-d1270'}, {'docid': u'R-d73'}, {'docid': u'R-d1271'}, {'docid': u'R-d71'}], 'qid': u'R-q2', 'runid': "LiLa’16" } R-q2 Q0 R-d70 1 0.9 LiLa16 R-q2 Q0 R-d72 2 0.8 LiLa16 R-q2 Q0 R-d74 3 0.7 LiLa16 R-q2 Q0 R-d75 4 0.6 LiLa16 R-q2 Q0 R-d1270 5 0.5 LiLa16 R-q2 Q0 R-d73 6 0.4 LiLa16 R-q2 Q0 R-d1271 7 0.3 LiLa16 R-q2 Q0 R-d71 8 0.2 LiLa16
Example { 'doclist': [ {'docid': u'R-d70'}, {'docid': u'R-d72'}, {'docid': u'R-d74'},
{'docid': u'R-d75'}, {'docid': u'R-d1270'}, {'docid': u'R-d73'}, {'docid': u'R-d1271'}, {'docid': u'R-d71'}], 'qid': u'R-q2', 'runid': "LiLa’16" } PUT http://api.living-labs.net/api/ participant/doclist/(key)/R-q2
Historical feedback • Go figure • http://doc.living-labs.net/en/latest/api- participant.html#historical-feedback
Exercise • Code for baseline (single field) retrieval and uploading
the ranking to the API is given in 02_ranker.py 1. Implement multi-field retrieval 2. Incorporate historical feedback into the ranking