That disconnect is why Cao turned to a technique called sentiment analysis \u2014 using AI to bring clarity to the tasting-note chaos for everyday wine drinkers. In a paper published<\/a> in the Harvard Data Science Review, Cao and her colleagues at SMU built an AI model that highlights specific words in a review that carry the positive or negative sentiment, the ones doing the persuading.<\/p>\n Sifting through the emotions<\/p>\n Breaking News<\/p>\n As humans, we do sentiment analysis all the time: while reading a text or a social media post or during real-life conversations. We judge whether the language conveys an upbeat, flat or ominous interaction. That intuitive read is sentiment analysis in miniature, Cao said.<\/p>\n \u201cNowadays, we live in a material world that\u2019s very buying and selling,\u201d Cao said. \u201cI don\u2019t think you will blindly purchase something \u2014 you look at the rating, you look at other customers\u2019 reviews or feedback.\u201d This process of mining others\u2019 opinions to make informed decisions is sentiment analysis, Cao added. \u201cIt\u2019s not a manual on how to use [a product] or how to drink it and what it tastes like. It\u2019s to see how enthusiastic people feel about it.\u201d<\/p>\n A Unitree G1 Edu humanoid robot capable of squatting, running, clapping, waving and lifting light objects greets visitors at the New York Stock Exchange in New York on Friday, August 1, 2025. <\/p>\n Ted Shaffrey \/ AP<\/p>\n But enthusiasm, and its absence, are felt keenly in the wine aisle. Reviews speak in vivid, bold descriptors about the aroma, body, flavor and finish that either beckon or warn off potential shoppers, said Carl Hudson, a certified wine specialist and educator with the Texas Wine Collective<\/a>.<\/p>\n A 2017 study<\/a> from the University of Adelaide in Australia found that evocative language not only influences which wine we buy but also shapes our emotions, nudging us to pay more for what we perceive as a more elaborate wine \u2014 and, in turn, to like the wine more when we finally sip it. <\/p>\n If words can steer wallets and palates, the obvious question is which words are doing the steering and how?<\/p>\n That question has long fascinated Cao, a self-described foodie interested in wine and consumer behavior. During the pandemic, with some grant funds left over, she and a few graduate students set out to answer that question. <\/p>\n A 2022 study<\/a> published in the Journal of Wine Economics showed that AI can scan wine reviews from Wine Spectator, a well-known wine magazine, and reliably predict a critic\u2019s score. Building on that idea, Cao and graduate student Chenyu Yang developed an AI system called an attention-based multiple instance model that analyzed over 140,000 Wine Spectator reviews written between 2005 and 2016.<\/p>\n If you imagine a wine review like a bag of grapes, the AI figures out which words are the grapes carrying the flavor, both individually and within a sentence.<\/p>\n Bottles of wine in a cooler at Neighborhood Cellar, a wine bar in Bishop Arts, on Thursday, Sept. 23, 2021, in Dallas. <\/p>\n Smiley N. Pool \/ Staff Photographer<\/p>\n Among the top positive sentiment words were ones anyone would expect: gorgeous, seductive, exquisite, velvet, sumptuous and impeccable. Others weren\u2019t \u2014 breed, soak, carpet and fabric \u2014 but in wine-speak, they read as praise.<\/p>\n Some words the AI flagged as negative sound otherwise positive in everyday speech: Picnic, watermelon, decent or breezy. Hudson said such words aren\u2019t desirable qualities, often signaling simplicity rather than depth. Picnic, for instance, suggests a white wine pleasant enough but short on elegance or character, he added. <\/p>\n Sentiment beyond wine<\/p>\n Compared with other sentiment-analysis models \u2014 like Google\u2019s BERT<\/a>, which was developed in 2018 and reads words in both directions to capture context \u2014 Cao and Yang\u2019s model hit 89.26% accuracy on the two-class test, edging past BERT\u2019s 89.12%. <\/p>\n Impressive as that is, especially against BERT, Cao said their AI isn\u2019t ready for the linguistic wilds of X or other online forums. On social media, a single word can swing from praise to snark depending on context; wine reviews are far more uniform. However, Cao is hoping to do more research and work on the AI to get it there. <\/p>\n Despite gloomy surveys about the future of AI<\/a> taking human jobs \u2014 and as its use creeps through the wine industry from farming sensors<\/a> to virtual sommeliers<\/a> \u2014 Hudson, a wine reviewer himself, welcomes the burgeoning technology. <\/p>\n \u201cIf people get something out of that, learn enough about wine to then accept the sentiment that AI gives them, or let the AI sentiment guide them to a particular type or set of wines,\u201d Hudson said, \u201cI think that\u2019s very valuable for our wine business. If we can grow more grapes and make more wine, I\u2019m all for that.\u201d <\/p>\n Beyond wine, Cao sees her AI helping decode the sentiment in other jargon-heavy worlds. An example could include doctor-speak during an annual check-up: translating it clearly could close communication gaps that contribute to health disparities<\/a>. The use of sentiment analysis in health care is being explored to some extent, such as using AI to sort whether patient reviews online of physicians<\/a> skew positive or negative.<\/p>\n \u201cNot only do I need to know what, but I would like to know how \u2014 why is this labeled in this way, why is this characterized in this way,\u201d Cao said. \u201cThat\u2019s the power of interpretable AI.\u201d<\/p>\n Miriam Fauzia is a science reporting fellow at The Dallas Morning News. Her fellowship is supported by the University of Texas at Dallas. The News makes all editorial decisions.<\/p>\n","protected":false},"excerpt":{"rendered":"You\u2019re standing in the wine aisle and Google a review for a bottle in your hand. Its description…\n","protected":false},"author":3,"featured_media":150632,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[5135],"tags":[5229,52560,1596,2105,50,18253,14971,358,3187,67,586,132,5230,48919,68,2969,65127],"class_list":{"0":"post-150631","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-dallas","8":"tag-america","9":"tag-bars-and-cocktails","10":"tag-dallas","11":"tag-food-and-drink","12":"tag-news","13":"tag-science-and-medicine","14":"tag-smu-mustangs","15":"tag-texas","16":"tag-tx","17":"tag-united-states","18":"tag-united-states-of-america","19":"tag-unitedstates","20":"tag-unitedstatesofamerica","21":"tag-university-park","22":"tag-us","23":"tag-usa","24":"tag-wine-spirits"},"share_on_mastodon":{"url":"https:\/\/pubeurope.com\/@us\/115038769976573193","error":""},"_links":{"self":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/150631","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/comments?post=150631"}],"version-history":[{"count":0,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/posts\/150631\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media\/150632"}],"wp:attachment":[{"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/media?parent=150631"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/categories?post=150631"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.europesays.com\/us\/wp-json\/wp\/v2\/tags?post=150631"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"A](\"https:\/\/www.europesays.com\/us\/wp-content\/uploads\/2025\/08\/N7E3JIEIEBDEXPYEIIUUNJWCDA.jpg\")
<\/p>\n![\"Bottles](\"https:\/\/www.europesays.com\/us\/wp-content\/uploads\/2025\/08\/HAIFTRABFRCYLJXGTAFWVH22HA.jpg\")
<\/p>\n