dataset and supplementary to "Framing Food Online Discourse: Employing Generative AI and Semantic Analysis for Digital Lexicography and Terminology Extraction"

Supplementary file Framing Food Online Discourse: Employing Generative AI and Semantic Analysis for Digital Lexicography and Terminology Extraction Malamatenia Panagiotou1,2*†, Konstantinos Gkatzionis1†, and Efstathios Kaloudis2† 1*Laboratory of Consumer and Sensory Perception of Food and Drinks, Department of Food Science and Nutrition, University of the Aegean, Metropolite Ioakeim 2, Myrina, 81400, Lemnos, Greece. 2Computer Simulation, Genomics and Data Analysis Laboratory, Department of Food Science and Nutrition, University of the Aegean, Leoforos Dimokratias 66, Myrina, 81400, Lemnos, Greece. *Corresponding author(s). E-mail(s): [email protected]; Contributing authors: [email protected]; stathiskaloudis Supplementary Table 1. Greeklish mappings as identified in the present study (the Greek letter /sound is followed by the Latin letter(s) that take its place during Greeklish transliteration). The pronunciation of the Greek letters/digraphs is also provided, as it affects the transliterating process. α (/a/) > a β (/v/) > b, v γ (/ɣ/, /ʝ/) > g δ (/ð/) > d, th ε (/e/) > e ζ (/z/) > z η (/i/) > i, e, h θ (/θ/) > th ι (/i/) > i κ (/k/) > k λ (/l/) > l μ (/m/) > m ν (/n/) > n ξ (/ks/) > ks, x ο (/o/) > o π (/p/) > p ρ (/r/) > r σ (/s/) > s τ (/t/) > t υ (/i/) > y, i φ (/f/) > f, ph χ (/h/, /ç/) > ch, h, x ψ (/ps/) > ps ω (/o/) > o αι (/e/) > ai, e ει (/i/) > ei, i οι (/i/) > oi, i αυ (/av/, /af/) > av, au, af ευ (/ev, ef/) > ev, eu, ef ου (/u/) > ou μπ (/b/) > b, mp ντ (/d/, /nd/) > d, nt γκ (/g/) > gg, gk, g τσ (/ts/) > ts τζ (/dz/) > tz Supplementary Table 2. Specific hashtags used to collect cheese related posts on Instagram. The Table shows all variations in Greeklish, Greek (correct and misspelled), and English (correct and misspelled). Greek local cheeses hashtags non-Greek cheeses hashtags PDO Ladotyri of Mytilene ladotyrimytilinis ladotyrimytilinis ladotyrimytilini ladotyri_cheese ladotyrimytilene ladotirimytilene ladotirimitilinis ladotirimitilini λαδοτύριμυτιλήνης λαδοτυριμυτιληνης λαδοτύριμυτιλίνης λαδοτύρι_μυτιλήνης λαδοτύρι_μυτιλλήνης Gouda κασεριgouda Mastelo® of Chios mastello tyrichiou mastelocheese μαστελο μαστέλο μαστελοχιου μαστελο_χιου μαστελο_χιωτικο Edam ενταμ ένταμ edam edamcheese PDO Kalathaki of Lemnos kalathakilimnou kalathakilemnos kalathakilemnou kalathaki_limnou kalathakilimnoscheese καλαθακιλημνου καλαθάκιλήμνου καλαθάκι_λήμνου Regatto ρεγκατο ρεγκάτο PDO Feta of Lemnos fetalimnou feta_limnou Mozzarella μοτσαρέλα μοτσαρελα μοτσαρέλατυρί motsarela Anthotyro of Lemnos anthotyrolimnou anthotirolimnou Emmental έμενταλ εμενταλ PDO Kasseri of Mytilene κασέρι_μυτιλήνης Parmesan παρμεζάνα παρμεζανα Kefalotyri of Mytilene κεφαλοτυριμυτιληνης kefalotyrimytilinis Blue cheese, Roquefort, Gorgonzola ροκφορ ροκφόρ γκοργκοντζόλα γκοργκοντζολα Melichloro of Lemnos melichlorolimnou melichloro_limnou melihloro melihlorocheeselemnosgreece melihlorocheese melixloro melixloro_limnou melichloro Cheddar τσενταρ τσένταρ μελίχλωρο μελιχλωρο μελιχλωροτυριλημνου μελιχλωρολημνου Kathoura of Icaria καθούρα καθουρα kathoura kathouracheese Cottage κότατζ κοτατζ (other hashtags searched for) anthotyros anthotyro_cheese ανθότυρο ανθοτυρο ανθότυρος ανθοτυρος limnoscheese lemnoscheese chioscheese samoscheese graviera γραβιέρα γραβιερα κεφαλοτυρι κεφαλοτύρι kefalotyri kefalotyricheese kefalotiri kefalotiricheese kefalotiri_cheese myzhthra myzithra myzithracheese myzithra_cheese mizithra mizithracheese mizithra_cheese μυζήθρα μυζηθρα μυζηθραξερη μυζήθραξηρή μυζήθρα_ξερή mytilinicheese τυριάμυτιλήνης cheese mytilini cheese mytilene cheese samos cheese ikaria cheese chios cheese xios cheese limnos cheese lemnos Supplementary Table 3. Example of the top 10 most frequent hashtags for the category CheeseVariety (excerpt). The full tables, including all categories for both local and non-Greek cheeses, and snails are provided in the online repository 1 [link: https://zenodo.org/records/17306360 ]. Local cheeses Non-Greek cheeses Hashtags Frequency Hashtags Frequency Graviera 6508 Mozzarella 540 Cheese 4374 Parmesan 532 Mizithra 3104 Gouda 231 Kefalotyri 2399 Cheddar 199 GreekCheese 1475 Cheese 175 Feta 1464 Feta 49 Anthotyro 684 Graviera 20 Manouri 327 Manouri 10 Halloumi 157 Ricotta 9 Gruyere 148 Anthotyro 8 Note: Normalized outcome (Greek, Greeklish, English variants merged under one canonical English form). Normalization mapping: Cheese Varieties • Graviera o Greek: γραβιέρα o Greeklish: graviera, graviéra, graviera_crete o English: graviera → Normalized form: Graviera • Feta o Greek: φέτα o Greeklish: feta, pheta o English: feta → Normalized form: Feta • Mizithra o Greek: μυζήθρα, μιζύθρα o Greeklish: mizithra, mizythra, mizithra_cheese o English: mizithra → Normalized form: Mizithra • Kefalotyri 1 Files provided in online repository: 1) Cheese_Hashtags_COMPARATIVE_FINAL.xlsx - Comparative (Local vs Non-Greek). Logic-first categorization. - CheeseVariety: ONLY actual cheeses; Greek/Greeklish/English variants merged to canonical English. - FoodPairing: dishes/ingredients (pizza, chicken, salad, bread); not treated as varieties. - README sheet explains the logic succinctly. 2) Snails_Hashtags_FINAL.xlsx - Logic-first categorization; NO normalization (hashtags preserved as collected). - README sheet; Counts_PerCategory (if available); All_Categorized reconstruction; verbatim percategory sheets. o Greek: κεφαλοτύρι o Greeklish: kefalotyri, kefalotiri o English: kefalotyri → Normalized form: Kefalotyri • Halloumi o Greek: χαλλούμι o Greeklish: halloumi, xaloumi o English: halloumi → Normalized form: Halloumi • Manouri o Greek: μανούρι o Greeklish: manouri o English: manouri → Normalized form: Manouri • Mozzarella o Greek: μοτσαρέλα, μοτσαρελα o Greeklish: mozarella, mozzarela, motzarella o English: mozzarella → Normalized form: Mozzarella • Parmesan o Greek: παρμεζάνα, παρμεζανα o Greeklish: parmezana, parmesana o English: parmesan → Normalized form: Parmesan • Gouda o Greek: γκούντα, γκουντα o Greeklish: gouda, gkounta o English: gouda → Normalized form: Gouda • Cheddar o Greek: τσένταρ, τσενταρ o Greeklish: tsentar, tsentar_cheese o English: cheddar → Normalized form: Cheddar Generic / Category Terms • Cheese o Greek: τυρί, τυρια o Greeklish: tyri, tiri o English: cheese → Normalized form: Cheese • GreekCheese o Greeklish: greekcheese, elliniko_tyri o English: greekcheese → Normalized form: GreekCheese Supplementary Table 4. TBX and OntoLex-lemon representations of “graviera”. TBX representation of “graviera” OntoLex-lemon representation of “graviera” <termEntry id="t_graviera"> <langSet xml:lang="en"> <tig> <term>graviera cheese</term> <descrip type="definition"> A hard yellow Greek cheese, traditionally made from sheep’s milk or a mixture of sheep and goat milk, recognized as a Protected Designation of Origin (PDO) product in Crete and Naxos. </descrip> <descrip type="frame">Consumption</descrip> <descrip type="frameElement">CheeseVariety</descrip> <descrip type="origin">Crete, Naxos</descrip> <descrip type="occasion">meze, dinner, holiday meal</descrip> <descrip type="foodPairing">wine, pasta, salad</descrip> <descrip type="assessment">authentic, traditional, tasty, PDO</descrip> <descrip type="emotion">nostalgia, pride, enjoyment</descrip> <admin type="status">preferredTerm</admin> <admin type="subjectField">food science, gastronomy</admin> </tig> </langSet> </termEntry> @prefix ontolex: <http://www.w3.org/ns/lemon/ontolex#> . @prefix lexinfo: <http://www.lexinfo.net/ontology/3.0/lexinfo#> . @prefix vartrans: <http://www.w3.org/ns/lemon/vartrans#> . @prefix lime: <http://www.w3.org/ns/lemon/lime#> . @prefix skos: <http://www.w3.org/2004/02/skos/core#> . @prefix dct: <http://purl.org/dc/terms/> . @prefix ex: <http://example.org/cheese/> . ######### Frame & frame-element style hooks (domain ontology layer) ex:ConsumptionFrame a ex:Frame ; skos:prefLabel "Consumption"@en . # FE-like properties used to encode roles ex:hasOrigin a owl:ObjectProperty . ex:hasOccasion a owl:ObjectProperty . ex:hasFoodPairing a owl:ObjectProperty . ex:hasAssessment a owl:ObjectProperty . ex:hasEmotion a owl:ObjectProperty . ex:hasVariety a owl:ObjectProperty . ######### Concept for Graviera (terminological/ontological node) ex:GRAVIERA a skos:Concept ; skos:prefLabel "Graviera (cheese)"@en ; skos:altLabel "γραβιέρα"@el ; skos:definition "Hard yellow Greek cheese, traditionally made from sheep’s milk or mixed sheep– goat milk; PDO in Crete and Naxos."@en ; dct:subject ex:ConsumptionFrame ; ex:hasVariety ex:CheeseVariety ; ex:hasOrigin ex:Crete , ex:Naxos ; ex:hasOccasion ex:Meze , ex:Dinner , ex:HolidayMeal ; ex:hasFoodPairing ex:Wine , ex:Pasta , ex:Salad ; ex:hasAssessment ex:Authentic , ex:Traditional , ex:Tasty , ex:PDO ; ex:hasEmotion ex:Nostalgia , ex:Pride , ex:Enjoyment . ######### Lexical entries (Greek & English) ex:graviera-el a ontolex:LexicalEntry ; lexinfo:partOfSpeech lexinfo:noun ; ontolex:canonicalForm [ ontolex:writtenRep "γραβιέρα"@el ] ; ontolex:sense ex:graviera-sense-el . ex:graviera-en a ontolex:LexicalEntry ; lexinfo:partOfSpeech lexinfo:noun ; ontolex:canonicalForm [ ontolex:writtenRep "graviera cheese"@en ] ; ontolex:otherForm [ ontolex:writtenRep "graviera"@en ] ; ontolex:sense ex:graviera-sense-en . # Senses point to the shared concept (denotation) ex:graviera-sense-el a ontolex:LexicalSense ; ontolex:reference ex:GRAVIERA . ex:graviera-sense-en a ontolex:LexicalSense ; ontolex:reference ex:GRAVIERA . ######### Cross-lingual link (vartrans module) ex:graviera-tr a vartrans:Translation ; vartrans:source ex:graviera-el ; vartrans:target ex:graviera-en ; vartrans:category vartrans:exact . ######### Simple individuals used by the concept (could be SKOS too) ex:CheeseVariety a skos:Concept ; skos:prefLabel "Cheese variety"@en . ex:Crete a skos:Concept ; skos:prefLabel "Crete"@en . ex:Naxos a skos:Concept ; skos:prefLabel "Naxos"@en . ex:Meze a skos:Concept ; skos:prefLabel "meze"@en . ex:Dinner a skos:Concept ; skos:prefLabel "dinner"@en . ex:HolidayMeal a skos:Concept ; skos:prefLabel "holiday meal"@en . ex:Wine a skos:Concept ; skos:prefLabel "wine"@en . ex:Pasta a skos:Concept ; skos:prefLabel "pasta"@en . ex:Salad a skos:Concept ; skos:prefLabel "salad"@en . ex:Authentic a skos:Concept ; skos:prefLabel "authentic"@en . ex:Traditional a skos:Concept ; skos:prefLabel "traditional"@en . ex:Tasty a skos:Concept ; skos:prefLabel "tasty"@en . ex:PDO a skos:Concept ; skos:prefLabel "PDO"@en . ex:Nostalgia a skos:Concept ; skos:prefLabel "nostalgia"@en . ex:Pride a skos:Concept ; skos:prefLabel "pride"@en . ex:Enjoyment a skos:Concept ; skos:prefLabel "enjoyment"@en . Note: Two LexicalEntry nodes (Greek/English) share one concept (ex:GRAVIERA) via ontolex:reference. Frame alignment: the concept is tagged with ex:ConsumptionFrame; FE-like relations are modeled as object properties (e.g., ex:hasOrigin). Cross-lingual alignment uses vartrans (vartrans:Translation with vartrans:exact). One can later align ex:* properties to community ontologies (e.g., FrameNet, FoodOn, GeoNames) if desired.