Exploration cycle finding a better dining experience: a framework of meal-plates

M
Matsushita LaboratoryMatsushita Laboratory
Exploration cycle
finding a better dining experience:
a framework of meal-plates
China Takahashia, Mitsunori Matsushitaa, Ryosuke Yamanishia
a Kansai University, Japan
27th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems
Athens, Greece, 06-08 September 2023
INTRODUCTION
• Dining is not just for nutrition
• One of the “experiential content” that enriches our daily lives
• The appeal of this dining experience is not only influenced by the
deliciousness of the meal, but also by the presentation of the meal
• e.g., the eccentricity of ingredients, cooking methods, serving and plates
• These balances are important to enhance the appeal of the dining experience
• We focused on the “plates”
1
• Plates are not just containers for serving meal
• If you are only looking for the role of putting meals in, one deep and
large plate should be enough
• However, there are various plates in the world
• We use it in various situations and in various meals
The selection of plates is one of the important factors contributing
to the enhancement of the dining experience
BACKGROUND
2
Wouldn't you like to eat the right meal every day?
3
• The selection of plates changes depending on the meal ingredients and
cooking method
• It is necessary to select according to the meal
• The current situation is…
• The user recalls his/her own dining experience and selects a plate
imaginatively
• Unable to improve because new discoveries and knowledge cannot
be obtained
PROBLEM (1)
4
• In many plates, there are multiple choices of meals that can be served
• It is necessary to select according to the preference and inspiration of
the person serving meals
• The current situation is…
• Users do not know their own preferences
We provide a system for supporting user's exploration
to find a suitable combination of meals and plates
PROBLEM (2)
5
IDEA: MEALS–PLATES EXPLORATION CYCLE
6
• This cycle is developed based on the conventional flow of plate selection
a plate
meals
A Part of plate selection
A Part of meal selection
Expand
Focused
Focused
Expand
a meal
plates
IDEA: MEALS–PLATES EXPLORATION CYCLE
7
a plate
meals
A Part of plate selection
A Part of meal selection
Expand
Focused
Focused
Expand
a meal
plates
computational support
A Part of plate selection
Recommend by
computer
Expand
Selection by user
Focused
Selection by user
Focused
Recommend by
computer
Expand
a plate
meals
A Part of meal selection
plates
a meal
Expand phase
Current situation
• New discoveries and knowledge
cannot be obtained
Our support
• Computer recommends multiple
meals(plates) from a single
plate(meal)
IDEA: MEALS–PLATES EXPLORATION CYCLE
8
A Part of plate selection
Recommend by
computer
Expand
Selection by user
Selection by user
Recommend by
computer
Expand
a plate
meals
A Part of meal selection
plates
a meal
Focused phase
Focused
Focused
Current situation
• Users do not know their own
preferences
Our support
• User selects a single plate(meal)
from multiple plates(meals) using
refine search
IDEA: MEALS–PLATES EXPLORATION CYCLE
9
A Part of plate selection
Recommend by
computer
Expand
Selection by user
Selection by user
Recommend by
computer
Expand
a plate
meals
A Part of meal selection
plates
a meal
Focused phase
Focused
Focused
• Users do not know their own
preferences
• User selects a single plate(meal)
from multiple meals(plates) using
filtering search
Not to explore individual plates or meals,
but to explore the plates by linking information about
the appropriate meal to each plate
Point of this idea
IDEA: MEALS–PLATES EXPLORATION CYCLE
10
THE GOAL OF THIS RESEARCH
11
• The user cyclically learns and selects the plate that matches the
meal and the meal that matches the plate
• The goal is to allow the user organize their own preferences and
select plates in an exploratory manner while understanding the
compatibility of meals with plates by myself.
Changes in users
Expand phase
(Computer recommends multiple meals(plates) from a single plate(meal))
• We need to identify the conditions that a meal suitable for the plate
should satisfy
• (A): converting meal information and plate information to
machine-readable data
• (B): associating meal information and plate information
converted to machine-readable data in (A)
HOW THE IDEA WORKS
12
• Data source
• Recipes from a cooking site (Cookpad[1])
• Define the meal elements involved in selecting a plate
• Ingredients
• Cooking behavior
(じゃがいも, にんじん, 玉ねぎ … 炒める, 茹でる, 切る)
(potato, carrot, onion … fry, boil, cut)
(1,1,1,1,0,・・・,1)
Represented by a binary vector
[1] Cookpad Inc., 2015. Cookpad dataset, the informatics research data repository, the national institute of informatics (dataset), https://doi.org/10.32130/idr.5.1.
(A-1) CONVERTING MEAL INFORMATION
13
• Data source
• E-commerce websites (Rakuten Ichiba Shopping Site[2])
• Define the plate elements involved in selecting a meal
• Size(the long side, the short side and height)→ measured value
• Shape (e.g., circles, corners, and flowers)→ binary
• Material (e.g., Japanese ceramics, lacquerware, and glass) → binary
(A-2) CONVERTING PLATE INFORMATION
14
[2] Rakuten Inc., 2014. Rakuten dataset, the informatics research data repository, the national institute of informatics, https://doi.org/10.32130/idr.2.0.
• How do you associate these two data?
• Associate by “meal name”, which is a common item of two data
• Cooking websites include recipe names and category names
• Product descriptions on E-commerce websites include descriptions of
examples of meals used
Pasta, curry, and Tenshinhan
(B) ASSOCIATING MEAL AND PLATE INFORMATION
15
• However, cooking websites and E-commerce websites do not have uniform
granularity of meal names
• meal names appeared in cooking websites are too fine
• e.g., Vegetable curry, indian curry, chicken curry…
• meal names appeared in the e-commerce website are too coarse
• e.g., curry
• Understanding the names of meal described in product descriptions on EC sites
• Obtained 117 new meal names not included in the meal category names on the cooking
site using CRF(Conditional Random Field)
• Hierarchical organization of meal names on recipe and e-commerce sites
(B) ASSOCIATING MEAL AND PLATE INFORMATION
16
• (A): converting meal information and plate information to machine-readable
data
• (B): associating meal information and plate information converted to machine-
readable data in (A)
• The following data are linked by the above (A) and (B)
• Ingredients
• Cooking behavior
meal information plate information
Meal name • Size 23
• Shape
• Material
LINKED RESULTS
17
Category: Curry
• Ingredients
• How to cook
Category: Stew
• Ingredients
• How to cook
Meal A
Meal B
This plate is perfect for serving pasta
Plate B
• Size
• Material
This plate is perfect for serving curry !
• Size
• Material
Plate A
18
• Shape
• Shape
EXAMPLE OF LINKED RESULTS
Category: Curry
• Ingredients
• How to cook
Category: Stew
• Ingredients
• How to cook
This plate is perfect for serving curry !
• Size
• Material
Meal A
Meal B
Plate A
This plate is perfect for serving pasta
Plate B
• Size
• Material
Corresponds plate information to meal information
by meal name (e.g., curry)
19
• Shape
• Shape
EXAMPLE OF LINKED RESULTS
Category: Curry
• Ingredients
• How to cook
Category: Stew
• Ingredients
• How to cook
Meal A
Meal B
This plate is perfect for serving curry !
• Size
• Material
Plate A
Applicable locations in the cycle
If the ingredients or how to cook of MealA and MealB
are similar, it is possible to serve MealB on PlateA
even if the category is not curry
20
• Shape
Category: Curry
• Ingredients
• How to cook
This plate is perfect for serving curry !
• Size
• Material
Meal A
Plate A
This plate is perfect for serving pasta
Plate B
• Size
• Material
Applicable locations in the cycle
If PlateA and PlateB are similar in size, shape and material,
it is possible to serve MealB on PlateA even if curry is not
mentioned in the product description
21
• Shape
• Shape
Focused phase
(User selects a single plate(meal) from multiple meals(plates) using filtering
search)
• Meal-focused phase
• Narrow down using conventional recipe recommendation technology
• e.g., meal similarity [3], ingredients [4], and preferences [5]
• Plate-focused phase
• Narrow down using the plate appearance characteristics
• e.g., color, pattern, shape, and size
HOW THE IDEA WORKS
22
[3] Wang, L., Li, Q., Li, N., Dong, G., Yang, Y., 2008. Substructure smilarity measurement in chinese recipes, in: Proc. 17th Int. Conf. on World Wide Web, pp. 979–988.
[4] Zhang, Q., Hu, R., Namee, B., Delany, S., 2008. Back to the future: Knowledge light case base cookery, in: Proc. 9th ECCBR, pp. 239––248
[5] Geleijnse, G.,Wang, L., Li, Q., 2010. Promoting tasty meals to support healthful eating, in:Wellness Informatics (WI)Workshop at CHI 2010.
• The selection of plates is one of the important factors contributing to
the enhancement of the dining experience
• we provide a system for supporting user's exploration to find a suitable
combination of meals and plates
Using Suggested Cycle,
• Users can organize their own preferences and select plates in an
exploratory manner while understanding the compatibility of meals
with plates
CONCLUSION
23
1 sur 24

Recommandé

copy-sample101 to foodservice par
copy-sample101 to foodservicecopy-sample101 to foodservice
copy-sample101 to foodservicenexchef
2.7K vues24 diapositives
HayateFukumoto_jsai2020 par
HayateFukumoto_jsai2020HayateFukumoto_jsai2020
HayateFukumoto_jsai2020Matsushita Laboratory
143 vues22 diapositives
Jsai2020 slide par
Jsai2020 slideJsai2020 slide
Jsai2020 slideMatsushita Laboratory
22 vues22 diapositives
Menu labeling risk mitigation par
Menu labeling  risk mitigationMenu labeling  risk mitigation
Menu labeling risk mitigationABC Research Laboratories
544 vues15 diapositives
Culinary R&D par
Culinary R&DCulinary R&D
Culinary R&DCatering By Michaels
729 vues45 diapositives
4th quarter cot par
4th quarter cot4th quarter cot
4th quarter cotJasmin Pionelo
42 vues48 diapositives

Contenu connexe

Similaire à Exploration cycle finding a better dining experience: a framework of meal-plates

Best practices Reducing Food Cost & Food Waste par
Best practices Reducing Food Cost & Food WasteBest practices Reducing Food Cost & Food Waste
Best practices Reducing Food Cost & Food WasteNevada Agriculture
2.5K vues27 diapositives
SCPC Exploratory PPI 042915 par
SCPC Exploratory PPI 042915SCPC Exploratory PPI 042915
SCPC Exploratory PPI 042915Steve Shutte
391 vues28 diapositives
Evaluation of popular diets par
Evaluation of popular dietsEvaluation of popular diets
Evaluation of popular dietsjohnsot4
534 vues10 diapositives
Chef Patrick's Pals, Nutrition and Culinary Program Webinar par
Chef Patrick's Pals, Nutrition and Culinary Program Webinar Chef Patrick's Pals, Nutrition and Culinary Program Webinar
Chef Patrick's Pals, Nutrition and Culinary Program Webinar Katie Baildon
567 vues30 diapositives
Dining preparation & Table Setting par
Dining preparation & Table SettingDining preparation & Table Setting
Dining preparation & Table SettingMerry Joy A. Longos
34.9K vues30 diapositives

Similaire à Exploration cycle finding a better dining experience: a framework of meal-plates(20)

Best practices Reducing Food Cost & Food Waste par Nevada Agriculture
Best practices Reducing Food Cost & Food WasteBest practices Reducing Food Cost & Food Waste
Best practices Reducing Food Cost & Food Waste
Nevada Agriculture2.5K vues
SCPC Exploratory PPI 042915 par Steve Shutte
SCPC Exploratory PPI 042915SCPC Exploratory PPI 042915
SCPC Exploratory PPI 042915
Steve Shutte391 vues
Evaluation of popular diets par johnsot4
Evaluation of popular dietsEvaluation of popular diets
Evaluation of popular diets
johnsot4534 vues
Chef Patrick's Pals, Nutrition and Culinary Program Webinar par Katie Baildon
Chef Patrick's Pals, Nutrition and Culinary Program Webinar Chef Patrick's Pals, Nutrition and Culinary Program Webinar
Chef Patrick's Pals, Nutrition and Culinary Program Webinar
Katie Baildon567 vues
Recommendation Architecture - OpenTable - RecSys 2014 - Large Scale Recommend... par Jeremy Schiff
Recommendation Architecture - OpenTable - RecSys 2014 - Large Scale Recommend...Recommendation Architecture - OpenTable - RecSys 2014 - Large Scale Recommend...
Recommendation Architecture - OpenTable - RecSys 2014 - Large Scale Recommend...
Jeremy Schiff2.5K vues
Research on Comparative Perception of Mess food vis a vis College Canteen food par SHAHBAAZ AHMED
Research on Comparative Perception of Mess food vis a vis College Canteen foodResearch on Comparative Perception of Mess food vis a vis College Canteen food
Research on Comparative Perception of Mess food vis a vis College Canteen food
SHAHBAAZ AHMED6.2K vues
Workshop: Cassava value chains comparison par CIAT
Workshop: Cassava value chains comparisonWorkshop: Cassava value chains comparison
Workshop: Cassava value chains comparison
CIAT1.1K vues
Grab and go trends how to enhance your offer 2017 par Rachael Sawtell
Grab and go trends   how to enhance your offer 2017Grab and go trends   how to enhance your offer 2017
Grab and go trends how to enhance your offer 2017
Rachael Sawtell505 vues
Ovenbot the real kitchen of the future par shawn212
Ovenbot the real kitchen of the futureOvenbot the real kitchen of the future
Ovenbot the real kitchen of the future
shawn212330 vues
Production of Nutritional Bars with Different Proportions of Oat Flour and Br... par asclepiuspdfs
Production of Nutritional Bars with Different Proportions of Oat Flour and Br...Production of Nutritional Bars with Different Proportions of Oat Flour and Br...
Production of Nutritional Bars with Different Proportions of Oat Flour and Br...
asclepiuspdfs45 vues
Assessing rice consumers'preferences and their willingness to pay in haiti ... par Cleeford PAVILUS
Assessing rice consumers'preferences and their willingness to pay in haiti   ...Assessing rice consumers'preferences and their willingness to pay in haiti   ...
Assessing rice consumers'preferences and their willingness to pay in haiti ...
Webinar: How to Use a Fishbone Diagram (Encore!) par GoLeanSixSigma.com
Webinar: How to Use a Fishbone Diagram (Encore!)Webinar: How to Use a Fishbone Diagram (Encore!)
Webinar: How to Use a Fishbone Diagram (Encore!)
GoLeanSixSigma.com4.1K vues
GraphTour: De-cyphering recipes with Neo4j par Neo4j
GraphTour: De-cyphering recipes with Neo4jGraphTour: De-cyphering recipes with Neo4j
GraphTour: De-cyphering recipes with Neo4j
Neo4j396 vues

Plus de Matsushita Laboratory

TaketoFujikawa_10thComicComputing2023 par
TaketoFujikawa_10thComicComputing2023TaketoFujikawa_10thComicComputing2023
TaketoFujikawa_10thComicComputing2023Matsushita Laboratory
93 vues24 diapositives
SayakaHayashi_FIT2023 par
SayakaHayashi_FIT2023SayakaHayashi_FIT2023
SayakaHayashi_FIT2023Matsushita Laboratory
17 vues19 diapositives
松下研究室紹介_関西大学高槻キャンパスオープンキャンパス par
松下研究室紹介_関西大学高槻キャンパスオープンキャンパス松下研究室紹介_関西大学高槻キャンパスオープンキャンパス
松下研究室紹介_関西大学高槻キャンパスオープンキャンパスMatsushita Laboratory
46 vues23 diapositives
ReonHata_JSAI2023 par
ReonHata_JSAI2023ReonHata_JSAI2023
ReonHata_JSAI2023Matsushita Laboratory
47 vues33 diapositives
HarukiShinkawa_FIT2023 par
HarukiShinkawa_FIT2023HarukiShinkawa_FIT2023
HarukiShinkawa_FIT2023Matsushita Laboratory
50 vues24 diapositives
TaketoFujikawa_KES2023 par
TaketoFujikawa_KES2023TaketoFujikawa_KES2023
TaketoFujikawa_KES2023Matsushita Laboratory
449 vues26 diapositives

Plus de Matsushita Laboratory(20)

松下研究室紹介_関西大学高槻キャンパスオープンキャンパス par Matsushita Laboratory
松下研究室紹介_関西大学高槻キャンパスオープンキャンパス松下研究室紹介_関西大学高槻キャンパスオープンキャンパス
松下研究室紹介_関西大学高槻キャンパスオープンキャンパス
Unification of Terminology for Accurate Communication among Experts --- Basic... par Matsushita Laboratory
Unification of Terminology for Accurate Communication among Experts --- Basic...Unification of Terminology for Accurate Communication among Experts --- Basic...
Unification of Terminology for Accurate Communication among Experts --- Basic...
触感に関わる共感覚的表現と基本6感情の対応関係の検証 par Matsushita Laboratory
触感に関わる共感覚的表現と基本6感情の対応関係の検証触感に関わる共感覚的表現と基本6感情の対応関係の検証
触感に関わる共感覚的表現と基本6感情の対応関係の検証
レシピの手順に着目した 複数の器特徴の推定 par Matsushita Laboratory
レシピの手順に着目した 複数の器特徴の推定レシピの手順に着目した 複数の器特徴の推定
レシピの手順に着目した 複数の器特徴の推定
複数の質感を複合的に提示可能な触覚提示デバイス par Matsushita Laboratory
複数の質感を複合的に提示可能な触覚提示デバイス複数の質感を複合的に提示可能な触覚提示デバイス
複数の質感を複合的に提示可能な触覚提示デバイス
効果音と抽象図形の動作の組み合わせによる印象変化に関する研究 par Matsushita Laboratory
効果音と抽象図形の動作の組み合わせによる印象変化に関する研究効果音と抽象図形の動作の組み合わせによる印象変化に関する研究
効果音と抽象図形の動作の組み合わせによる印象変化に関する研究
携帯端末を用いたポインティングによる室内空間でのアドホックな情報アクセス手法に関する研究 par Matsushita Laboratory
携帯端末を用いたポインティングによる室内空間でのアドホックな情報アクセス手法に関する研究携帯端末を用いたポインティングによる室内空間でのアドホックな情報アクセス手法に関する研究
携帯端末を用いたポインティングによる室内空間でのアドホックな情報アクセス手法に関する研究
Kokogatari:実環境を介したリレー小説執筆ツール par Matsushita Laboratory
Kokogatari:実環境を介したリレー小説執筆ツールKokogatari:実環境を介したリレー小説執筆ツール
Kokogatari:実環境を介したリレー小説執筆ツール
Visualization of the Relationship Between Lectures and Laboratories Using SSNMF par Matsushita Laboratory
Visualization of the Relationship Between Lectures and Laboratories Using SSNMFVisualization of the Relationship Between Lectures and Laboratories Using SSNMF
Visualization of the Relationship Between Lectures and Laboratories Using SSNMF

Dernier

Transcript: The Details of Description Techniques tips and tangents on altern... par
Transcript: The Details of Description Techniques tips and tangents on altern...Transcript: The Details of Description Techniques tips and tangents on altern...
Transcript: The Details of Description Techniques tips and tangents on altern...BookNet Canada
130 vues15 diapositives
Top 10 Strategic Technologies in 2024: AI and Automation par
Top 10 Strategic Technologies in 2024: AI and AutomationTop 10 Strategic Technologies in 2024: AI and Automation
Top 10 Strategic Technologies in 2024: AI and AutomationAutomationEdge Technologies
14 vues14 diapositives
RADIUS-Omnichannel Interaction System par
RADIUS-Omnichannel Interaction SystemRADIUS-Omnichannel Interaction System
RADIUS-Omnichannel Interaction SystemRADIUS
15 vues21 diapositives
AI: mind, matter, meaning, metaphors, being, becoming, life values par
AI: mind, matter, meaning, metaphors, being, becoming, life valuesAI: mind, matter, meaning, metaphors, being, becoming, life values
AI: mind, matter, meaning, metaphors, being, becoming, life valuesTwain Liu 刘秋艳
35 vues16 diapositives
Igniting Next Level Productivity with AI-Infused Data Integration Workflows par
Igniting Next Level Productivity with AI-Infused Data Integration Workflows Igniting Next Level Productivity with AI-Infused Data Integration Workflows
Igniting Next Level Productivity with AI-Infused Data Integration Workflows Safe Software
225 vues86 diapositives
Tunable Laser (1).pptx par
Tunable Laser (1).pptxTunable Laser (1).pptx
Tunable Laser (1).pptxHajira Mahmood
23 vues37 diapositives

Dernier(20)

Transcript: The Details of Description Techniques tips and tangents on altern... par BookNet Canada
Transcript: The Details of Description Techniques tips and tangents on altern...Transcript: The Details of Description Techniques tips and tangents on altern...
Transcript: The Details of Description Techniques tips and tangents on altern...
BookNet Canada130 vues
RADIUS-Omnichannel Interaction System par RADIUS
RADIUS-Omnichannel Interaction SystemRADIUS-Omnichannel Interaction System
RADIUS-Omnichannel Interaction System
RADIUS15 vues
AI: mind, matter, meaning, metaphors, being, becoming, life values par Twain Liu 刘秋艳
AI: mind, matter, meaning, metaphors, being, becoming, life valuesAI: mind, matter, meaning, metaphors, being, becoming, life values
AI: mind, matter, meaning, metaphors, being, becoming, life values
Igniting Next Level Productivity with AI-Infused Data Integration Workflows par Safe Software
Igniting Next Level Productivity with AI-Infused Data Integration Workflows Igniting Next Level Productivity with AI-Infused Data Integration Workflows
Igniting Next Level Productivity with AI-Infused Data Integration Workflows
Safe Software225 vues
Architecting CX Measurement Frameworks and Ensuring CX Metrics are fit for Pu... par NUS-ISS
Architecting CX Measurement Frameworks and Ensuring CX Metrics are fit for Pu...Architecting CX Measurement Frameworks and Ensuring CX Metrics are fit for Pu...
Architecting CX Measurement Frameworks and Ensuring CX Metrics are fit for Pu...
NUS-ISS37 vues
Voice Logger - Telephony Integration Solution at Aegis par Nirmal Sharma
Voice Logger - Telephony Integration Solution at AegisVoice Logger - Telephony Integration Solution at Aegis
Voice Logger - Telephony Integration Solution at Aegis
Nirmal Sharma17 vues
.conf Go 2023 - How KPN drives Customer Satisfaction on IPTV par Splunk
.conf Go 2023 - How KPN drives Customer Satisfaction on IPTV.conf Go 2023 - How KPN drives Customer Satisfaction on IPTV
.conf Go 2023 - How KPN drives Customer Satisfaction on IPTV
Splunk88 vues
Business Analyst Series 2023 - Week 3 Session 5 par DianaGray10
Business Analyst Series 2023 -  Week 3 Session 5Business Analyst Series 2023 -  Week 3 Session 5
Business Analyst Series 2023 - Week 3 Session 5
DianaGray10209 vues
Web Dev - 1 PPT.pdf par gdsczhcet
Web Dev - 1 PPT.pdfWeb Dev - 1 PPT.pdf
Web Dev - 1 PPT.pdf
gdsczhcet55 vues
Future of Learning - Khoong Chan Meng par NUS-ISS
Future of Learning - Khoong Chan MengFuture of Learning - Khoong Chan Meng
Future of Learning - Khoong Chan Meng
NUS-ISS33 vues
Emerging & Future Technology - How to Prepare for the Next 10 Years of Radica... par NUS-ISS
Emerging & Future Technology - How to Prepare for the Next 10 Years of Radica...Emerging & Future Technology - How to Prepare for the Next 10 Years of Radica...
Emerging & Future Technology - How to Prepare for the Next 10 Years of Radica...
NUS-ISS16 vues

Exploration cycle finding a better dining experience: a framework of meal-plates

  • 1. Exploration cycle finding a better dining experience: a framework of meal-plates China Takahashia, Mitsunori Matsushitaa, Ryosuke Yamanishia a Kansai University, Japan 27th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems Athens, Greece, 06-08 September 2023
  • 2. INTRODUCTION • Dining is not just for nutrition • One of the “experiential content” that enriches our daily lives • The appeal of this dining experience is not only influenced by the deliciousness of the meal, but also by the presentation of the meal • e.g., the eccentricity of ingredients, cooking methods, serving and plates • These balances are important to enhance the appeal of the dining experience • We focused on the “plates” 1
  • 3. • Plates are not just containers for serving meal • If you are only looking for the role of putting meals in, one deep and large plate should be enough • However, there are various plates in the world • We use it in various situations and in various meals The selection of plates is one of the important factors contributing to the enhancement of the dining experience BACKGROUND 2
  • 4. Wouldn't you like to eat the right meal every day? 3
  • 5. • The selection of plates changes depending on the meal ingredients and cooking method • It is necessary to select according to the meal • The current situation is… • The user recalls his/her own dining experience and selects a plate imaginatively • Unable to improve because new discoveries and knowledge cannot be obtained PROBLEM (1) 4
  • 6. • In many plates, there are multiple choices of meals that can be served • It is necessary to select according to the preference and inspiration of the person serving meals • The current situation is… • Users do not know their own preferences We provide a system for supporting user's exploration to find a suitable combination of meals and plates PROBLEM (2) 5
  • 7. IDEA: MEALS–PLATES EXPLORATION CYCLE 6 • This cycle is developed based on the conventional flow of plate selection a plate meals A Part of plate selection A Part of meal selection Expand Focused Focused Expand a meal plates
  • 8. IDEA: MEALS–PLATES EXPLORATION CYCLE 7 a plate meals A Part of plate selection A Part of meal selection Expand Focused Focused Expand a meal plates computational support
  • 9. A Part of plate selection Recommend by computer Expand Selection by user Focused Selection by user Focused Recommend by computer Expand a plate meals A Part of meal selection plates a meal Expand phase Current situation • New discoveries and knowledge cannot be obtained Our support • Computer recommends multiple meals(plates) from a single plate(meal) IDEA: MEALS–PLATES EXPLORATION CYCLE 8
  • 10. A Part of plate selection Recommend by computer Expand Selection by user Selection by user Recommend by computer Expand a plate meals A Part of meal selection plates a meal Focused phase Focused Focused Current situation • Users do not know their own preferences Our support • User selects a single plate(meal) from multiple plates(meals) using refine search IDEA: MEALS–PLATES EXPLORATION CYCLE 9
  • 11. A Part of plate selection Recommend by computer Expand Selection by user Selection by user Recommend by computer Expand a plate meals A Part of meal selection plates a meal Focused phase Focused Focused • Users do not know their own preferences • User selects a single plate(meal) from multiple meals(plates) using filtering search Not to explore individual plates or meals, but to explore the plates by linking information about the appropriate meal to each plate Point of this idea IDEA: MEALS–PLATES EXPLORATION CYCLE 10
  • 12. THE GOAL OF THIS RESEARCH 11 • The user cyclically learns and selects the plate that matches the meal and the meal that matches the plate • The goal is to allow the user organize their own preferences and select plates in an exploratory manner while understanding the compatibility of meals with plates by myself. Changes in users
  • 13. Expand phase (Computer recommends multiple meals(plates) from a single plate(meal)) • We need to identify the conditions that a meal suitable for the plate should satisfy • (A): converting meal information and plate information to machine-readable data • (B): associating meal information and plate information converted to machine-readable data in (A) HOW THE IDEA WORKS 12
  • 14. • Data source • Recipes from a cooking site (Cookpad[1]) • Define the meal elements involved in selecting a plate • Ingredients • Cooking behavior (じゃがいも, にんじん, 玉ねぎ … 炒める, 茹でる, 切る) (potato, carrot, onion … fry, boil, cut) (1,1,1,1,0,・・・,1) Represented by a binary vector [1] Cookpad Inc., 2015. Cookpad dataset, the informatics research data repository, the national institute of informatics (dataset), https://doi.org/10.32130/idr.5.1. (A-1) CONVERTING MEAL INFORMATION 13
  • 15. • Data source • E-commerce websites (Rakuten Ichiba Shopping Site[2]) • Define the plate elements involved in selecting a meal • Size(the long side, the short side and height)→ measured value • Shape (e.g., circles, corners, and flowers)→ binary • Material (e.g., Japanese ceramics, lacquerware, and glass) → binary (A-2) CONVERTING PLATE INFORMATION 14 [2] Rakuten Inc., 2014. Rakuten dataset, the informatics research data repository, the national institute of informatics, https://doi.org/10.32130/idr.2.0.
  • 16. • How do you associate these two data? • Associate by “meal name”, which is a common item of two data • Cooking websites include recipe names and category names • Product descriptions on E-commerce websites include descriptions of examples of meals used Pasta, curry, and Tenshinhan (B) ASSOCIATING MEAL AND PLATE INFORMATION 15
  • 17. • However, cooking websites and E-commerce websites do not have uniform granularity of meal names • meal names appeared in cooking websites are too fine • e.g., Vegetable curry, indian curry, chicken curry… • meal names appeared in the e-commerce website are too coarse • e.g., curry • Understanding the names of meal described in product descriptions on EC sites • Obtained 117 new meal names not included in the meal category names on the cooking site using CRF(Conditional Random Field) • Hierarchical organization of meal names on recipe and e-commerce sites (B) ASSOCIATING MEAL AND PLATE INFORMATION 16
  • 18. • (A): converting meal information and plate information to machine-readable data • (B): associating meal information and plate information converted to machine- readable data in (A) • The following data are linked by the above (A) and (B) • Ingredients • Cooking behavior meal information plate information Meal name • Size 23 • Shape • Material LINKED RESULTS 17
  • 19. Category: Curry • Ingredients • How to cook Category: Stew • Ingredients • How to cook Meal A Meal B This plate is perfect for serving pasta Plate B • Size • Material This plate is perfect for serving curry ! • Size • Material Plate A 18 • Shape • Shape EXAMPLE OF LINKED RESULTS
  • 20. Category: Curry • Ingredients • How to cook Category: Stew • Ingredients • How to cook This plate is perfect for serving curry ! • Size • Material Meal A Meal B Plate A This plate is perfect for serving pasta Plate B • Size • Material Corresponds plate information to meal information by meal name (e.g., curry) 19 • Shape • Shape EXAMPLE OF LINKED RESULTS
  • 21. Category: Curry • Ingredients • How to cook Category: Stew • Ingredients • How to cook Meal A Meal B This plate is perfect for serving curry ! • Size • Material Plate A Applicable locations in the cycle If the ingredients or how to cook of MealA and MealB are similar, it is possible to serve MealB on PlateA even if the category is not curry 20 • Shape
  • 22. Category: Curry • Ingredients • How to cook This plate is perfect for serving curry ! • Size • Material Meal A Plate A This plate is perfect for serving pasta Plate B • Size • Material Applicable locations in the cycle If PlateA and PlateB are similar in size, shape and material, it is possible to serve MealB on PlateA even if curry is not mentioned in the product description 21 • Shape • Shape
  • 23. Focused phase (User selects a single plate(meal) from multiple meals(plates) using filtering search) • Meal-focused phase • Narrow down using conventional recipe recommendation technology • e.g., meal similarity [3], ingredients [4], and preferences [5] • Plate-focused phase • Narrow down using the plate appearance characteristics • e.g., color, pattern, shape, and size HOW THE IDEA WORKS 22 [3] Wang, L., Li, Q., Li, N., Dong, G., Yang, Y., 2008. Substructure smilarity measurement in chinese recipes, in: Proc. 17th Int. Conf. on World Wide Web, pp. 979–988. [4] Zhang, Q., Hu, R., Namee, B., Delany, S., 2008. Back to the future: Knowledge light case base cookery, in: Proc. 9th ECCBR, pp. 239––248 [5] Geleijnse, G.,Wang, L., Li, Q., 2010. Promoting tasty meals to support healthful eating, in:Wellness Informatics (WI)Workshop at CHI 2010.
  • 24. • The selection of plates is one of the important factors contributing to the enhancement of the dining experience • we provide a system for supporting user's exploration to find a suitable combination of meals and plates Using Suggested Cycle, • Users can organize their own preferences and select plates in an exploratory manner while understanding the compatibility of meals with plates CONCLUSION 23