diff --git a/build.xml b/build.xml index 6d7fb00..529bdda 100644 --- a/build.xml +++ b/build.xml @@ -43,12 +43,11 @@ - + - + - - + + diff --git a/test/dkohl/bayes/example/builders/FoodExampleBuilder.java b/src/dkohl/bayes/builders/FoodExampleBuilder.java similarity index 99% rename from test/dkohl/bayes/example/builders/FoodExampleBuilder.java rename to src/dkohl/bayes/builders/FoodExampleBuilder.java index 3bcc9c8..7ee532f 100644 --- a/test/dkohl/bayes/example/builders/FoodExampleBuilder.java +++ b/src/dkohl/bayes/builders/FoodExampleBuilder.java @@ -1,4 +1,4 @@ -package dkohl.bayes.example.builders; +package dkohl.bayes.builders; import java.util.HashSet; import java.util.LinkedList; diff --git a/src/dkohl/bayes/builders/FoodNetBuilder.java b/src/dkohl/bayes/builders/FoodNetBuilder.java index 9bc22ad..a04a2bf 100644 --- a/src/dkohl/bayes/builders/FoodNetBuilder.java +++ b/src/dkohl/bayes/builders/FoodNetBuilder.java @@ -10,7 +10,6 @@ import net.woodyfolsom.cs6601.p2.RecipeBook; import net.woodyfolsom.cs6601.p2.Survey; import dkohl.bayes.bayesnet.BayesNet; import dkohl.bayes.estimation.MaximumLikelihoodEstimation; -import dkohl.bayes.example.builders.FoodExampleBuilder; import dkohl.bayes.probability.Assignment; import dkohl.bayes.probability.Probability; import dkohl.bayes.probability.Variable; diff --git a/test/dkohl/bayes/example/FoodExampleTest.java b/test/dkohl/bayes/example/FoodExampleTest.java index bbdc069..73879c0 100644 --- a/test/dkohl/bayes/example/FoodExampleTest.java +++ b/test/dkohl/bayes/example/FoodExampleTest.java @@ -9,7 +9,7 @@ import java.util.List; import org.junit.Test; import dkohl.bayes.bayesnet.BayesNet; -import dkohl.bayes.example.builders.FoodExampleBuilder; +import dkohl.bayes.builders.FoodExampleBuilder; import dkohl.bayes.inference.EnumerateAll; import dkohl.bayes.probability.ProbabilityAssignment; import dkohl.bayes.probability.Variable; diff --git a/writeup/P2 Proposal.aux b/writeup/P2 Proposal.aux index 35a856e..7ced2ae 100644 --- a/writeup/P2 Proposal.aux +++ b/writeup/P2 Proposal.aux @@ -3,14 +3,19 @@ \citation{corney} \citation{janzenxiang} \citation{truyen} +\citation{truyen} \citation{murphy} +\citation{murphy} +\citation{russelnorvig} +\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces Our Bayesian net modeling user preferences. The top layer describes the categories Meat and Vegetable. We have a control variable vegetarian for Meat, such that it will always evaluate to $0$ when there is meat involved in a dish and we have a vegetarian diner. The mid layer describes the preference for different ingredients. The last layer is a Gaussian predicting the users preferences.}}{2}} +\newlabel{img:bayes_net}{{1}{2}} \bibstyle{plain} \bibdata{p2refs} \bibcite{corney}{1} \bibcite{janzenxiang}{2} \bibcite{murphy}{3} -\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces Our Baysian net modeling user preferences}}{2}} -\newlabel{img:bayes_net}{{1}{2}} -\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Estimated vs. Actual Survey Dish Ratings}}{2}} -\bibcite{truyen}{4} -\bibcite{vanboekel}{5} +\bibcite{russelnorvig}{4} +\bibcite{truyen}{5} +\bibcite{vanboekel}{6} +\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces Estimated vs. surveyed dish ratings}}{3}} +\newlabel{result}{{2}{3}} diff --git a/writeup/P2 Proposal.bbl b/writeup/P2 Proposal.bbl index 2092b90..c999c61 100644 --- a/writeup/P2 Proposal.bbl +++ b/writeup/P2 Proposal.bbl @@ -1,34 +1,39 @@ -\begin{thebibliography}{1} - -\bibitem{corney} -David Corney. -\newblock {Designing Food} with {Bayesian Belief Networks}. -\newblock {\em ACDM}, 2000. - -\bibitem{janzenxiang} -M.~Janzen and Y.~Xiang. -\newblock {Probabilistic Reasoning} in {Meal Planning} in {Intelligent - Fridges}. -\newblock {\em 16th Conference of the Canadian Society for Computational - Studies of Intelligence}, 2003. - -\bibitem{murphy} -Kevin Murphy. -\newblock A {Brief Introduction} to {Graphical Models} and {Bayesian Networks}. -\newblock http://www.cs.ubc.ca/~murphyk/Bayes/bnintro.html, 1998. -\newblock [Online; accessed 22-Feb-2012]. - -\bibitem{truyen} -Tran~The Truyen, Dinh~Q. Phung, and Svetha Venkatesh. -\newblock Preference {Networks}: {Probabilistic} {Models} for {Recommendation} - {Systems}. -\newblock {\em Proceedings of the {Sixth Australasian Conference} on {Data - Mining}}, 2007. - -\bibitem{vanboekel} -Stein~A. van Boekel, M.A.J.S. and A.H.C. van Bruggen. -\newblock {Bayesian Statistics} and {Quality Modelling} in the {Agro-food - Production Chain}. -\newblock {\em Proceedings of the Frontis workshop}, 2004. - -\end{thebibliography} +\begin{thebibliography}{1} + +\bibitem{corney} +David Corney. +\newblock {Designing Food} with {Bayesian Belief Networks}. +\newblock {\em ACDM}, 2000. + +\bibitem{janzenxiang} +M.~Janzen and Y.~Xiang. +\newblock {Probabilistic Reasoning} in {Meal Planning} in {Intelligent + Fridges}. +\newblock {\em 16th Conference of the Canadian Society for Computational + Studies of Intelligence}, 2003. + +\bibitem{murphy} +Kevin Murphy. +\newblock A {Brief Introduction} to {Graphical Models} and {Bayesian Networks}. +\newblock http://www.cs.ubc.ca/~murphyk/Bayes/bnintro.html, 1998. +\newblock [Online; accessed 22-Feb-2012]. + +\bibitem{russelnorvig} +S.~Russel and P.~Norvig. +\newblock {\em Artificial Intelligence; A Modern Approach}. +\newblock Prentice Hall, third edition, 2010. + +\bibitem{truyen} +Tran~The Truyen, Dinh~Q. Phung, and Svetha Venkatesh. +\newblock Preference {Networks}: {Probabilistic} {Models} for {Recommendation} + {Systems}. +\newblock {\em Proceedings of the {Sixth Australasian Conference} on {Data + Mining}}, 2007. + +\bibitem{vanboekel} +Stein~A. van Boekel, M.A.J.S. and A.H.C. van Bruggen. +\newblock {Bayesian Statistics} and {Quality Modelling} in the {Agro-food + Production Chain}. +\newblock {\em Proceedings of the Frontis workshop}, 3, 2004. + +\end{thebibliography} diff --git a/writeup/P2 Proposal.blg b/writeup/P2 Proposal.blg index 81f4791..aa8e381 100644 --- a/writeup/P2 Proposal.blg +++ b/writeup/P2 Proposal.blg @@ -1,28 +1,51 @@ -This is BibTeX, Version 0.99dThe top-level auxiliary file: P2 Proposal.aux -The style file: plain.bst -Database file #1: p2refs.bib -I was expecting a `,' or a `}'---line 14 of file p2refs.bib - : - : CITY = {Wageningen} -(Error may have been on previous line) -I'm skipping whatever remains of this entry -I was expecting a `,' or a `}'---line 23 of file p2refs.bib - : - : CITY = {London} -(Error may have been on previous line) -I'm skipping whatever remains of this entry -I was expecting a `,' or a `}'---line 31 of file p2refs.bib - : - : CITY = {London} -(Error may have been on previous line) -I'm skipping whatever remains of this entry -I was expecting a `,' or a `}'---line 40 of file p2refs.bib - : - : CITY = {Guelph} -(Error may have been on previous line) -I'm skipping whatever remains of this entry -Too many commas in name 1 of "van Boekel, M.A.J.S., Stein, A. and van Bruggen, A.H.C." for entry vanboekel -while executing---line 1049 of file plain.bst -Too many commas in name 1 of "van Boekel, M.A.J.S., Stein, A. and van Bruggen, A.H.C." for entry vanboekel -while executing---line 1090 of file plain.bst -(There were 6 error messages) +This is BibTeX, Version 0.99d (TeX Live 2010) +Capacity: max_strings=35307, hash_size=35307, hash_prime=30011 +The top-level auxiliary file: P2 Proposal.aux +The style file: plain.bst +Database file #1: p2refs.bib +Too many commas in name 1 of "van Boekel, M.A.J.S., Stein, A. and van Bruggen, A.H.C." for entry vanboekel +while executing---line 1049 of file plain.bst +Too many commas in name 1 of "van Boekel, M.A.J.S., Stein, A. and van Bruggen, A.H.C." for entry vanboekel +while executing---line 1090 of file plain.bst +You've used 6 entries, + 2118 wiz_defined-function locations, + 529 strings with 4926 characters, +and the built_in function-call counts, 1457 in all, are: += -- 138 +> -- 71 +< -- 0 ++ -- 29 +- -- 22 +* -- 72 +:= -- 272 +add.period$ -- 19 +call.type$ -- 6 +change.case$ -- 29 +chr.to.int$ -- 0 +cite$ -- 6 +duplicate$ -- 59 +empty$ -- 113 +format.name$ -- 22 +if$ -- 295 +int.to.chr$ -- 0 +int.to.str$ -- 6 +missing$ -- 6 +newline$ -- 34 +num.names$ -- 12 +pop$ -- 34 +preamble$ -- 1 +purify$ -- 23 +quote$ -- 0 +skip$ -- 44 +stack$ -- 0 +substring$ -- 31 +swap$ -- 6 +text.length$ -- 0 +text.prefix$ -- 0 +top$ -- 0 +type$ -- 22 +warning$ -- 0 +while$ -- 12 +width$ -- 7 +write$ -- 66 +(There were 2 error messages) diff --git a/writeup/P2 Proposal.log b/writeup/P2 Proposal.log index 08d34cd..561d382 100644 --- a/writeup/P2 Proposal.log +++ b/writeup/P2 Proposal.log @@ -1,4 +1,4 @@ -This is pdfTeX, Version 3.1415926-2.3-1.40.12 (MiKTeX 2.9) (preloaded format=pdflatex 2012.1.11) 12 MAR 2012 22:10 +This is pdfTeX, Version 3.1415926-2.3-1.40.12 (MiKTeX 2.9) (preloaded format=pdflatex 2012.1.11) 13 MAR 2012 07:08 entering extended mode **D:/workspace/cs6601p2/writeup/P2*Proposal.tex ("D:/workspace/cs6601p2/writeup/P2 Proposal.tex" @@ -46,13 +46,41 @@ Package: titlesec 2011/12/15 v2.10.0 Sectioning titles \titlewidthlast=\dimen106 \titlewidthfirst=\dimen107 ) -("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\graphics\graphicx.sty" -Package: graphicx 1999/02/16 v1.0f Enhanced LaTeX Graphics (DPC,SPQR) +("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\geometry\geometry.sty" +Package: geometry 2010/09/12 v5.6 Page Geometry ("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\graphics\keyval.sty" Package: keyval 1999/03/16 v1.13 key=value parser (DPC) \KV@toks@=\toks15 ) +("C:\Program Files (x86)\MiKTeX 2.9\tex\generic\oberdiek\ifpdf.sty" +Package: ifpdf 2011/01/30 v2.3 Provides the ifpdf switch (HO) +Package ifpdf Info: pdfTeX in PDF mode is detected. +) +("C:\Program Files (x86)\MiKTeX 2.9\tex\generic\oberdiek\ifvtex.sty" +Package: ifvtex 2010/03/01 v1.5 Switches for detecting VTeX and its modes (HO) +Package ifvtex Info: VTeX not detected. +) +("C:\Program Files (x86)\MiKTeX 2.9\tex\generic\ifxetex\ifxetex.sty" +Package: ifxetex 2010/09/12 v0.6 Provides ifxetex conditional +) +\Gm@cnth=\count87 +\Gm@cntv=\count88 +\c@Gm@tempcnt=\count89 +\Gm@bindingoffset=\dimen108 +\Gm@wd@mp=\dimen109 +\Gm@odd@mp=\dimen110 +\Gm@even@mp=\dimen111 +\Gm@layoutwidth=\dimen112 +\Gm@layoutheight=\dimen113 +\Gm@layouthoffset=\dimen114 +\Gm@layoutvoffset=\dimen115 +\Gm@dimlist=\toks16 + +("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\geometry\geometry.cfg")) +("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\graphics\graphicx.sty" +Package: graphicx 1999/02/16 v1.0f Enhanced LaTeX Graphics (DPC,SPQR) + ("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\graphics\graphics.sty" Package: graphics 2009/02/05 v1.0o Standard LaTeX Graphics (DPC,SPQR) @@ -73,10 +101,10 @@ Package: infwarerr 2010/04/08 v1.3 Providing info/warning/message (HO) ("C:\Program Files (x86)\MiKTeX 2.9\tex\generic\oberdiek\ltxcmds.sty" Package: ltxcmds 2011/04/18 v1.20 LaTeX kernel commands for general use (HO) ) -\Gread@gobject=\count87 +\Gread@gobject=\count90 )) -\Gin@req@height=\dimen108 -\Gin@req@width=\dimen109 +\Gin@req@height=\dimen116 +\Gin@req@width=\dimen117 ) ("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\ams\math\amsmath.sty" Package: amsmath 2000/07/18 v2.13 AMS math features @@ -88,51 +116,51 @@ Package: amstext 2000/06/29 v2.01 ("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\ams\math\amsgen.sty" File: amsgen.sty 1999/11/30 v2.0 -\@emptytoks=\toks16 -\ex@=\dimen110 +\@emptytoks=\toks17 +\ex@=\dimen118 )) ("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\ams\math\amsbsy.sty" Package: amsbsy 1999/11/29 v1.2d -\pmbraise@=\dimen111 +\pmbraise@=\dimen119 ) ("C:\Program Files (x86)\MiKTeX 2.9\tex\latex\ams\math\amsopn.sty" Package: amsopn 1999/12/14 v2.01 operator names ) -\inf@bad=\count88 +\inf@bad=\count91 LaTeX Info: Redefining \frac on input line 211. -\uproot@=\count89 -\leftroot@=\count90 +\uproot@=\count92 +\leftroot@=\count93 LaTeX Info: Redefining \overline on input line 307. -\classnum@=\count91 -\DOTSCASE@=\count92 +\classnum@=\count94 +\DOTSCASE@=\count95 LaTeX Info: Redefining \ldots on input line 379. LaTeX Info: Redefining \dots on input line 382. LaTeX Info: Redefining \cdots on input line 467. \Mathstrutbox@=\box27 \strutbox@=\box28 -\big@size=\dimen112 +\big@size=\dimen120 LaTeX Font Info: Redeclaring font encoding OML on input line 567. LaTeX Font Info: Redeclaring font encoding OMS on input line 568. -\macc@depth=\count93 -\c@MaxMatrixCols=\count94 +\macc@depth=\count96 +\c@MaxMatrixCols=\count97 \dotsspace@=\muskip10 -\c@parentequation=\count95 -\dspbrk@lvl=\count96 -\tag@help=\toks17 -\row@=\count97 -\column@=\count98 -\maxfields@=\count99 -\andhelp@=\toks18 -\eqnshift@=\dimen113 -\alignsep@=\dimen114 -\tagshift@=\dimen115 -\tagwidth@=\dimen116 -\totwidth@=\dimen117 -\lineht@=\dimen118 -\@envbody=\toks19 +\c@parentequation=\count98 +\dspbrk@lvl=\count99 +\tag@help=\toks18 +\row@=\count100 +\column@=\count101 +\maxfields@=\count102 +\andhelp@=\toks19 +\eqnshift@=\dimen121 +\alignsep@=\dimen122 +\tagshift@=\dimen123 +\tagwidth@=\dimen124 +\totwidth@=\dimen125 +\lineht@=\dimen126 +\@envbody=\toks20 \multlinegap=\skip48 \multlinetaggap=\skip49 -\mathdisplay@stack=\toks20 +\mathdisplay@stack=\toks21 LaTeX Info: Redefining \[ on input line 2666. LaTeX Info: Redefining \] on input line 2667. ) @@ -154,50 +182,94 @@ LaTeX Font Info: ... okay on input line 11. LaTeX Font Info: Checking defaults for U/cmr/m/n on input line 11. LaTeX Font Info: ... okay on input line 11. +*geometry* driver: auto-detecting +*geometry* detected driver: pdftex +*geometry* verbose mode - [ preamble ] result: +* driver: pdftex +* paper: +* layout: +* layoutoffset:(h,v)=(0.0pt,0.0pt) +* modes: +* h-part:(L,W,R)=(50.58878pt, 513.11745pt, 50.58878pt) +* v-part:(T,H,B)=(50.58878pt, 693.79243pt, 50.58878pt) +* \paperwidth=614.295pt +* \paperheight=794.96999pt +* \textwidth=513.11745pt +* \textheight=693.79243pt +* \oddsidemargin=-21.68121pt +* \evensidemargin=-21.68121pt +* \topmargin=-21.68121pt +* \headheight=0.0pt +* \headsep=0.0pt +* \topskip=10.0pt +* \footskip=30.0pt +* \marginparwidth=4.0pt +* \marginparsep=10.0pt +* \columnsep=22.58437pt +* \skip\footins=9.0pt plus 4.0pt minus 2.0pt +* \hoffset=0.0pt +* \voffset=0.0pt +* \mag=1000 +* \@twocolumntrue +* \@twosidefalse +* \@mparswitchfalse +* \@reversemarginfalse +* (1in=72.27pt=25.4mm, 1cm=28.453pt) + ("C:\Program Files (x86)\MiKTeX 2.9\tex\context\base\supp-pdf.mkii" [Loading MPS to PDF converter (version 2006.09.02).] -\scratchcounter=\count100 -\scratchdimen=\dimen119 +\scratchcounter=\count103 +\scratchdimen=\dimen127 \scratchbox=\box29 -\nofMPsegments=\count101 -\nofMParguments=\count102 -\everyMPshowfont=\toks21 -\MPscratchCnt=\count103 -\MPscratchDim=\dimen120 -\MPnumerator=\count104 -\makeMPintoPDFobject=\count105 -\everyMPtoPDFconversion=\toks22 +\nofMPsegments=\count104 +\nofMParguments=\count105 +\everyMPshowfont=\toks22 +\MPscratchCnt=\count106 +\MPscratchDim=\dimen128 +\MPnumerator=\count107 +\makeMPintoPDFobject=\count108 +\everyMPtoPDFconversion=\toks23 ) -Underfull \hbox (badness 4254) in paragraph at lines 67--70 -[] \OT1/cmr/bx/n/10 Data ac-cui-si-tion \OT1/cmr/m/n/10 We first ac-cu-mu-lated +Underfull \hbox (badness 8094) in paragraph at lines 89--94 +[] \OT1/cmr/bx/n/10 Data ac-qui-si-tion \OT1/cmr/m/n/10 We first ac-cu-mu-lated a di- [] - -File: bayes.png Graphic file (type png) - -Package pdftex.def Info: bayes.png used on input line 99. -(pdftex.def) Requested size: 237.13594pt x 115.43013pt. - [1{C:/ProgramData/MiKTeX/2.9/pdftex/config/pdftex.map} + +Underfull \hbox (badness 1881) in paragraph at lines 89--94 +\OT1/cmr/m/n/10 verse col-lec-tion of license-free sam-ple recipes from + [] + + +Underfull \hbox (badness 2600) in paragraph at lines 89--94 +\OT1/cmr/m/n/10 web sites such as \OT1/cmr/m/it/10 Dark-star's Meal-Master Reci +pes + [] + +[1{C:/ProgramData/MiKTeX/2.9/pdftex/config/pdftex.map} ] + +File: bayes.png Graphic file (type png) + +Package pdftex.def Info: bayes.png used on input line 117. +(pdftex.def) Requested size: 220.73839pt x 115.39198pt. + Package amsmath Warning: Foreign command \over; (amsmath) \frac or \genfrac should be used instead -(amsmath) on input line 133. +(amsmath) on input line 161. - -LaTeX Warning: Reference `rms-table' on page 2 undefined on input line 150. - - +[2 ] + File: BayesChefChart.png Graphic file (type png) -Package pdftex.def Info: BayesChefChart.png used on input line 154. -(pdftex.def) Requested size: 248.42812pt x 182.55777pt. +Package pdftex.def Info: BayesChefChart.png used on input line 254. +(pdftex.def) Requested size: 256.55872pt x 188.53769pt. -Overfull \hbox (11.29218pt too wide) in paragraph at lines 154--155 +Overfull \hbox (11.29218pt too wide) in paragraph at lines 254--255 [][] [] @@ -216,36 +288,35 @@ Underfull \hbox (badness 10000) in paragraph at lines 16--20 \OT1/cmr/m/n/10 works. http://www.cs.ubc.ca/ mur- [] -[2 ]) [3 -] ("D:\workspace\cs6601p2\writeup\P2 Proposal.aux") +Underfull \hbox (badness 1038) in paragraph at lines 34--38 +[]\OT1/cmr/m/n/10 Stein A. van Boekel, M.A.J.S. and A.H.C. van + [] -LaTeX Warning: There were undefined references. - - ) +) [3 ] +("D:\workspace\cs6601p2\writeup\P2 Proposal.aux") ) Here is how much of TeX's memory you used: - 2023 strings out of 494045 - 25481 string characters out of 3145969 - 87176 words of memory out of 3000000 - 5325 multiletter control sequences out of 15000+200000 + 2521 strings out of 494045 + 32956 string characters out of 3145969 + 104365 words of memory out of 3000000 + 5807 multiletter control sequences out of 15000+200000 11026 words of font info for 36 fonts, out of 3000000 for 9000 715 hyphenation exceptions out of 8191 - 27i,8n,28p,716b,219s stack positions out of 5000i,500n,10000p,200000b,50000s -{C:/Program Files (x86)/MiKTeX 2.9/fonts/enc/dvips/fontname/8r.enc} -Output written on "P2 Proposal.pdf" (3 pages, 190917 bytes). + 27i,8n,32p,572b,220s stack positions out of 5000i,500n,10000p,200000b,50000s +{C:/Program Files (x86)/MiKT +eX 2.9/fonts/enc/dvips/fontname/8r.enc} +Output written on "P2 Proposal.pdf" (3 pages, 187866 bytes). PDF statistics: - 55 PDF objects out of 1000 (max. 8388607) + 51 PDF objects out of 1000 (max. 8388607) 0 named destinations out of 1000 (max. 500000) 11 words of extra memory for PDF output out of 10000 (max. 10000000) diff --git a/writeup/P2 Proposal.pdf b/writeup/P2 Proposal.pdf index 8ba717c..4992b73 100644 Binary files a/writeup/P2 Proposal.pdf and b/writeup/P2 Proposal.pdf differ diff --git a/writeup/P2 Proposal.tex b/writeup/P2 Proposal.tex index 30fe380..ca03e4e 100644 --- a/writeup/P2 Proposal.tex +++ b/writeup/P2 Proposal.tex @@ -2,7 +2,7 @@ \usepackage{latex8} \usepackage{titlesec} -% \usepackage[margin=0.5in]{geometry} +\usepackage[margin=0.7in]{geometry} \usepackage{graphicx} \usepackage{amsmath} @@ -27,31 +27,53 @@ allergies and religious beliefs. Many professional catering services handle this problem by allowing guests to select from a very limited menu. We introduce a dish recommendation system based on Bayesian Networks modeling user preferences. -We predict the meals from a data base of recepices that most likely match the varied tastes +We predict the meals from a data base of recipes that most likely match the varied tastes of the customers, using a limited set of ingredients. This type of expert system -would be of great use to a catering service or restaurant which needs to rapdily decide on +would be of great use to a catering service or restaurant which needs to rapidly decide on a small number of dishes which would be acceptable for a large dinner party, -given diverse requirements and preferences. +given diverse requirements and preferences. + +\paragraph*{Bayesian Catering: A use case.} +Imagine that you run a catering service and have to +plan an event with a customer. You can +create a variety of dishes and now you +want to discuss with your clients which one to +serve. In order to get a better idea of which preferences +and needs you clients will have, you let them fill +out a survey in advance, where they rate a small amount of +your dishes on a scale from $1 - 10$ and inform you about +hard constraints like allergies, religious constraints +or vegetarians. You then use those results in order to predict +the ratings for the rest of your dishes and present the clients +the top $k$ results. If such a system works this will save time and will lead +to a better customer satisfaction since you can present them dishes +they will most probably like but still surprise them (since you have +not presented them what the already rated). After the dinner, +participants could rate the dishes served at the party which would iteratively improve the process for future customers. \section*{Related Work} Boekel and Corney propose using Bayesian Networks to model consumer needs in food production chains \cite{vanboekel} \cite{corney}. Janzen and Xiang propose an intelligent refrigerator capable of generating meal plans based on inventory -and past food choices \cite{janzenxiang}. Baysian networks have also been +and past food choices \cite{janzenxiang}. +We suggest that these approaches are limited in that they +only consider the preferences of a single (or supposed 'typical') +user rather than a group. Baysian networks have also been applied to recommendation systems before in on-line social networks \cite{truyen} making predictions of the form ``if you bought those items what is the probability you would like to -buy that''. We suggest that these approaches are limited in that they -only consider the preferences of a single (or supposed 'typical') user rather than a group. +buy that''. This method also uses bayesian networks +for prediction and our approach is similar or inspired by +some of Truyens \cite{truyen}. \section*{Approach} The approached problem is to pick a single meal which best meets the requirements and tastes of different people dining together. We learn a predictive -baysian net from a survey distributed to participants of the meal as +Bayesian net from a survey distributed to participants of the meal as training data in order to capture their preferences. The dishes -in the questionaire are selected such that all ingrediants +in the questionnaire are selected such that all ingredients are covered. The participants rate each dish on a scale from one to ten and give additional information like vegetarians. For new dishes we then predict the maximum likelihood @@ -63,100 +85,199 @@ end how we trained the modeled net from gathered data and predicted the value for different recipes. -\paragraph*{Data accuisition} -We first accumulated a diverse collection of license-free sample recipes from web sites such as \emph{Darkstar's Meal-Master Recipes} (http://home.earthlink.net/~darkstar105/). Next, we converted these recipes from flat text files to well-formed XML using the Krecipes application for Debian Linux. Finally, we created a representative data set representing several diners' preference for +\paragraph*{Data acquisition} +We first accumulated a diverse collection of license-free sample recipes from web sites such as \emph{Darkstar's Meal-Master Recipes} (http://home.earthlink.net/~darkstar105/). Next, we converted these recipes from flat text files to well-formed XML using the `Krecipes' application for Debian Linux. Finally, we created a representative data set representing several diners' preference for 24 of these recipes, using a simple survey of the type 'rate on a -scale of 1 to 10, 10 being favorite and 1 being least favorite'. Furthermore, users were allowed to specify a vegetarian or nut-free meal preference. +scale of 1 to 10, 10 being favorite and 1 being least favorite'. +Furthermore, users were allowed to specify a vegetarian +or nut-free meal preference. -%daniel is here \paragraph*{Knowledge Engineering} We model the diners' various taste preferences using -a Bayes net. We model the taste - -\begin{description} -\item[Layer 1] The first layer models a general preference towards -different food categories like vegetables or meat. -As one can see, the food categories are dependent -on the general meal preference. For example -being vegetarian will exclude meat and will -support vegetables. - -\item[Layer 2] Specific flavors and ingredients. Each ingredient is conditioned -by the food category to which it belongs. -\end{description} - -If we need to model hard constraints, like - - +a Bayes net. The net consists of three node types. +We call them ``control nodes'', ``taste nodes'' +and ``rating nodes''. A ``preference node'' +models the probability of a diners' preference towards an +ingredient ($P(likes tomato)$, $P(likes potato)$) or a category +($P(likes meat)$). These variables are discrete. The ingredients +are conditional independent from each other but conditioned +by the food category they belong to (see Figure \ref{img:bayes_net} +the two top layers). A control node can definitely reject a dish, +by evaluating to $0$ in certain conditions. For +example if someone is vegetarian and the presented dish contains +meet, the control variable for vegetarian will evaluate to $0$ and +so the probability for the whole dish will become $0$. +So the vegetarian variable is conditioned by meat. +The third type in the net is a preference node, it is continuous +and models the dish rating given a set of ingredients. The overall net is shown in Figure \ref{img:bayes_net}. -Given a recipe with a list of ingredients $I = i_1,...,i_n$ -and a Bayesian network capturing user preferences -we can calculate the probability of users liking the dish given -the probabilities of liking each ingrediant. -\begin{figure} +\begin{figure}[ht] \centering -\includegraphics[width=\linewidth]{bayes} -\caption{Our Baysian net modeling user preferences} +\includegraphics[width=0.9\linewidth]{bayes} +\caption{Our Bayesian net modeling user preferences. The top layer + describes the categories Meat and Vegetable. We have a control + variable vegetarian for Meat, such that it will always evaluate to + $0$ when there is meat involved in a dish and we have a vegetarian + diner. The mid layer describes the preference for different + ingredients. The last layer is a Gaussian predicting the users preferences.} \label{img:bayes_net} \end{figure} %\subsection*{implementation} -\paragraph*{Learning and Predicting} +\paragraph*{Learning user preferences} In order to estimate the model parameters, the system will be trained with statistics about taste and preferences given a set of dishes with ratings -from multiple users. From that information we can directly calculate -the probabilities for the ingredients using Maximum Likelihood Learning \cite{murphy}. +from multiple users. The training set is generated from +the questionnaires we distributed. +An example for a survey output could look like this (Ingredients, Rating): ``Pork, +Potatoes, 8''. In order to perform normal Maximum Likelihood Learning +\cite{murphy} we have to have information about all variables +(``Pork, Potatoes, Tomatoes, Beef, Meat, Vegetables, Rating''). +We perform several steps in order to transform from the survey input +to a training instance. +First we discretize the values such that all given variables (in our +case Pork and Potatoes) are set to ``true'' if the value is above +a certain threshold (in our experiments $5$) and ``false'' otherwise. +In that way ``liking things rated $>$ 5'' appear more often in the +training set and will be assigned with a higher probability. +We add categories by including the category of each ingredient +from the survey. If the ingredient is liked, the category is too +and if it is not, the category is not liked too. The last step +is to add all values that are not in the recipe as ``false'' +to the training instance. - -%\subsection*{Meal Optimization} -In order to model food preferences, we implemented -a custom Baysian net library in java with minimal use of third party libraries (e.g. for XML input). -We chose to implement our own Library, for maximum flexibility and to ensure that the learning algorithm functions precisely as follows: - -The library -uses the sum-product algorithm for -inference and maximum likelihood learning -for parameter estimation. In our implementation -we support discrete as well as continous -probability distributions. Discrete distributions -can be modeled as tables or as trees. -In our implementation only continous distributions with discrete parents -are supported. A continous distribution is then modeled as a mapping -of all possible combination of it' s parents to a gaussian. -Given a data set, the parameters of a discrete variable $X$ are -estimated as +From a set of those preprocessed assignments, we can directly calculate +the probabilities for the ingredients using Maximum Likelihood +Learning \cite{murphy}. For example for an assignment of a +conditional variable $P(X = x \mid Y_1 = y_1, ... Y_2 = y2)$, +we count how often we observe the configuration $X = x, Y_1 = y_1, +... Y_2 = y2$ and how often we count $Y_1 = y_1, ... Y_2 = y2$ +in our data set. The maximum likelihood is then defined as \begin{align} P(X = x| Y_1 = y_1, ... Y_2 = y2) =\\ N(X = x| Y_1 = y_1, ... Y_2 = y2) \over N(Y_1 = y_1, ... Y_2 = y2) \end{align} where $N(A)$ is the number of times event $A$occurs in the data set. +For a continuous variable like rating, we estimate a Gaussian for +each combination of it's parents. For example if the rating variable +is dependent on beef and tomatoes, we would estimate 4 Gaussians, +one for each possible combination of beef and tomatoes. So during +training we would estimate mean and variance for all cases where +$(tomato = true, potato = true)$, $(tomato = false, potato = true)$ +and so on. + +\paragraph*{Inferring maximum likelihood rating} +Having estimated the probabilities of such a net, we can infer +the maximum likelihood rating of a unseen dish while observing only +a set of ingredients. Therefore, we iterate over all possible +ratings ($1 - 10$) and compute the probability of this rating. +The maximum probability is the maximum likelihood rating +for that dish. We use the \emph{enumerateAll} algorithm \cite{russelnorvig}, +for the probability calculations. + +\paragraph*{Implementation} +In order to model food preferences, we implemented +a custom Bayesian net library in Java with +minimal use of third party libraries (e.g. for XML input). +We chose to implement our own Library, +for maximum flexibility and to ensure that the learning algorithm +functions precisely as follows: +The library uses the sum-product algorithm for +inference and maximum likelihood learning +for parameter estimation. In our implementation +we support discrete as well as continuous +probability distributions. Discrete distributions +can be modeled as tables or as trees. +In our implementation only continuous distributions with discrete parents +are supported. A continuous distribution is then modeled as a mapping +of all possible combination of its parents to a Gaussian set. \section*{Evaluation} -The application model will be trained using a sparse subset (50\%) of the survey data and the optimization problem solved for the inferred constraints. As shown below, the calculated preferences for recipes which were not used to train the Bayes net are quite close to the actual survey data, which essentially reflects the following preferences (Sample ratings are on a 1-10 scale): +In an experiment we collected $24$ ratings from $4$ persons. +We trained the Bayes net using a sparse subset (50\%) of the survey data. Then we evaluated the rest of the +recipes (which are all unseen) and calculated +the maximum likelihood rating. +As shown below, the calculated preferences for recipes +which were not used to train the Bayes net are quite close to the +actual survey data, which essentially reflects the following +preferences +(Sample ratings are on a 1-10 scale): \begin{description} -\item[Diner 1] No allergies, prefers all dishes equally (5) - +\item[Diner 1] Prefers all dishes equally (5) \item[Diner 2] Vegetarian, meat dishes are (1), remainder are (9) - -\item[Diner 3] Nut Alleregy, prefers meat (6) to vegetarian (4) to desert (3) - -\item[Diner 4] No allergies, prefers Pork and Desserts (9), remainder are (3) - +\item[Diner 3] Prefers meat (6) to vegetarian (4) to desert (3) +\item[Diner 4] Prefers Pork and Desserts (9), remainder are (3) \end{description} -Next, we calculate the correlation between the application's ranking of all dishes and the actual ranking as determined by the user surveys. We suggest that a high degree of correlation indicates that the system has the potential to accurately appraise constrained group food preferences for dishes which are not part of the survey, given sufficiently detailed recipe information. As \ref{rms-table} shows, the estimated food preferences are quite close to the actual mean ratings over all diners for the dishes which were not used to train the Bayes net. The root mean-square-error for calculated vs. surveyed meal preferences is approximately 1.0. +Next, we calculate the error between the application's ranking of all +dishes and the actual ranking as determined by the user surveys. We +suggest that a low error indicates that the system has the potential +to accurately appraise constrained group food preferences for dishes +which are not part of the survey, given sufficiently detailed recipe +information. As the Table and Figure \ref{result} show, the estimated food preferences are quite close to the actual mean ratings over all diners for the dishes which were not used to train the Bayes net. The root mean-square-error for calculated vs. surveyed meal preferences is approximately 1.92. -\begin{figure}[h!] +\begin{table}[ht] +\begin{tabular}{ | l | l | l | } +\hline +Dish & Est. & Actual Avg.\\ \hline +Southwest Smoothie: & &\\ +DAIRY & 5 & 5.5 \\ \hline +Bayou Shrimp Creole: & &\\ +TOMATO & 9 & 3.75 \\ \hline +Crab Burgers: & &\\ +EGGS & 5 & 3.75 \\ \hline +Broiled Flounder: & &\\ +GENERIC NUTS, EGGS & 5 & 3.75 \\ \hline +Baked Steak And Lima Beans: & &\\ +TOMATO, SUGAR & 2 & 3.75 \\ \hline +Eggplant Lasagna: & &\\ +GLUTEN & 5 & 5.25 \\ \hline +Salisbury Steak: & &\\ +GLUTEN, DAIRY, BEEF & 6 & 3.75 \\ \hline +Meatless Loaf: & &\\ +SPICE & 5 & 5.25 \\ \hline +Lemon Pork Chops: & &\\ +PORK, SUGAR & 5 & 5.25 \\ \hline +Fava Bean Burgers: & &\\ +EGGS, POTATO & 3 & 5.25 \\ \hline +Angel Hair Pesto Primavera: & &\\ +GENERIC NUTS, SPICE & 5 & 5.25 \\ \hline +%\hline +\end{tabular} +\end{table} + +\begin{figure}[ht] \centering \includegraphics[width=0.5 \textwidth]{BayesChefChart.png} -\caption{Estimated vs. Actual Survey Dish Ratings} +\caption{Estimated vs. surveyed dish ratings} +\label{result} \end{figure} -Note the outlier at Dish \#2 (Bayou Shrimp Creole). The strong preference for this dish is a result of the ingredient list containing primarily shrimp and tomato. Unlike beef and pork, the seafood category was not implemented in the knowledge enginerring of the net. Consequently, this dish is incorrectly deemed to be vegetarian-compatible. The same issue had previously occurred at Dish \#5 (Baked Steak and Lima Beans) until 'steak' was added to the recipe parser as a synonym for beef, and therefore a type of meat. +Note the outlier at Dish \#2 (Bayou Shrimp Creole). The strong preference for this dish is a result of the ingredient list containing primarily shrimp and tomato. Unlike beef and pork, the seafood category was not implemented in the knowledge engineering of the net. Consequently, this dish is incorrectly deemed to be vegetarian-compatible. The same issue had previously occurred at Dish \#5 (Baked Steak and Lima Beans) until 'steak' was added to the recipe parser as a synonym for beef, and therefore a type of meat. +\section*{Conclusion} +We proposed, implemented and evaluated a food preference prediction system that +is capable of predicting how much a user would like a new, unseen recipes. +We discussed how to encode user preference towards ingredients and +categories in a Bayes Net and how to add control variables in order +to exclude dishes that users have to avoid, such as meat in the case of +vegetarians. Furthermore, we presented our learning +scheme for such a Bayes net using data from a small survey +and how to predict the user rating for unseen dishes. +In an evaluation we showed that the net can predict +preferences when learned from a sparse data set. +So in a real life setting, where people plan +a dinner with a catering service, +a few participants could rate a small amount of recipes +in an on-line service +and the system could actually predict the scores +on the rest of the caterers data base. +The top $k$ with the answers highest predicted +rating of the system could be used to assemble the final +dinner. \bibliographystyle{plain} \bibliography{p2refs} diff --git a/writeup/bayes.png b/writeup/bayes.png index 7d13160..28de665 100644 Binary files a/writeup/bayes.png and b/writeup/bayes.png differ diff --git a/writeup/p2refs.bib b/writeup/p2refs.bib index d9cf557..2f2df1a 100644 --- a/writeup/p2refs.bib +++ b/writeup/p2refs.bib @@ -10,8 +10,8 @@ AUTHOR = {van Boekel, M.A.J.S., Stein, A. and van Bruggen, A.H.C.}, TITLE = {{Bayesian Statistics} and {Quality Modelling} in the {Agro-food Production Chain}}, YEAR = {2004}, - JOURNAL = {Proceedings of the Frontis workshop} - CITY = {Wageningen} + JOURNAL = {Proceedings of the Frontis workshop}, + CITY = {Wageningen}, VOLUME = {3} } @@ -19,7 +19,7 @@ AUTHOR = {David Corney}, TITLE = {{Designing Food} with {Bayesian Belief Networks}}, YEAR = {2000}, - JOURNAL = {ACDM} + JOURNAL = {ACDM}, CITY = {London} } @@ -27,7 +27,7 @@ AUTHOR = {Tran The Truyen and Dinh Q. Phung and Svetha Venkatesh}, TITLE = {Preference {Networks}: {Probabilistic} {Models} for {Recommendation} {Systems}}, YEAR = {2007}, - JOURNAL = {Proceedings of the {Sixth Australasian Conference} on {Data Mining}} + JOURNAL = {Proceedings of the {Sixth Australasian Conference} on {Data Mining}}, CITY = {London} } @@ -36,7 +36,7 @@ AUTHOR = {Janzen, M. and Xiang, Y.}, TITLE = {{Probabilistic Reasoning} in {Meal Planning} in {Intelligent Fridges}}, YEAR = {2003}, - JOURNAL = {16th Conference of the Canadian Society for Computational Studies of Intelligence} + JOURNAL = {16th Conference of the Canadian Society for Computational Studies of Intelligence}, CITY = {Guelph} }