The Beer Mood Matcher

Introduction

Have you ever wondered how a simple sip of beer can transport you through a large panel of emotions and moods? Welcome to the fascinating world of beer reviews, where each review is not just about the taste but an emotional journey!

In every beer review, there lies a hidden tale of emotions and moods, waiting to be unraveled. From the bold adventurer seeking the thrill of a new flavour to the nostalgic soul reminiscing a bygone era with a classic stout, each review is a testament to how beer evokes more than just taste buds.

How do the characteristics of different beers align with the emotions identified in reviews?

Every beer tells a story, evoking a spectrum of emotions in its drinker. By delving into the rich textual content of the reviews, we can find the perfect beer, that matches our desired mood. Our research is focused on determining how different beers relate to various moods expressed in reviews on popular platforms of BeerAdvocate and RateBeer. Reviews are analyzed and scored based on primary emotions including joy, sadness, anger, fear, love, and surprise applying NLP techniques, specifically Sentence-Bert for semantic similarity detection. Furthermore, we will determine the relationship between the characteristics of various beers and the moods they evoke, as identified in user reviews, which can benefit the marketing industry. Moreover, the connections between the emotions users exhibit while writing reviews and their various ratings are examined. Finally, a mood-based recommendation list of beer styles is created based on the mapped emotion scores.

Firstly, Let's get an overview of the dataset!

We have two datasets of beer reviews, details as below:

RateBeer

BeerAdvocate

Data pre-processing

For data processing, we first removed duplicates from the RateBeer and BeerAdvocate datasets and then eliminated rows with missing 'text' values. Next, we analyzed the review counts to understand distribution patterns, noting that many beers had fewer than two reviews. To ensure robust emotion analysis, we filtered out beers with less than two reviews, retaining only those with sufficient review data. This approach streamlined our datasets for more accurate and reliable emotion analysis.

Wordcloud for RB

Wordcloud for BA

RateBeer's cloud highlights "taste" and subtle characteristics like "bit," with additional focus on aroma, texture, and aftertaste through words like "nose," "mouthfeel," and "finish." The BeerAdvocate word cloud emphasizes "head," "malt," and "taste," focusing on foam quality and flavour, while "body" and color terms are less frequent.

Emotion Score Determination

Emotion scores for the RateBeer and BeerAdvocate reviews datasets were computed through a  process that leveraged NLP to utilise the power of machine learning. We converted the textual reviews into useful numerical data by utilising the SentenceTransformer library, and then we encoded the data into semantic vectors. We've done a preliminary analysis of the various reviews and decided to use a model named 'paraphrase-multilingual-MiniLM-L12-v2' that was competent at capturing the variations across languages. We measured the similarity scores for each review by comparing these vectors with the embeddings for key emotions:
 joy, sadness, anger, fear, love, and surprise.

The distribution of the Emotion Scores

RateBeer

BeerAdvocate

The chart shows that the emotion scores for the two websites under analysis are distributed between -0.2 and 0.8. The feelings of surprise and joy have higher average scores, indicating that these are prevalent emotions in beer reviews. On the other hand, fear has the lowest average, suggesting that it is not as frequently connected to the experience of analysing beer. 

The low standard deviations found in all of the emotion scores point to consistency in emotional reactions, providing the reliability of our data in expressing the emotions associated with each type of beer. This consistency makes it possible for us to apply these findings to individual beer experiences and to confidently link particular emotional profiles with particular beer styles.

The correlation in-between Emotion Scores

We see that the correlation between the analyzed emotions is generally high (which can seem reasonable given they all express an emotion, leading to high similarity), yet there are distinct similarities and differences for websites. 

RateBeer

BeerAdvocate

Firstly, it is possible to say that the correlation between the emotions in comments on the RateBeer site is less compared to the BeerAdvocate site. The lowest similarity between joy and fear emotions on the RateBeer website, coming in at 0.22, indicates that our model distinguishes well between comments that contain emotions of fear and those of joy. The second lowest correlation score, with 0.29, is between joy and anger emotions similar to fear, suggesting that there are some distinctions for some of the opposite emotions on the RB website.

On the other hand, on both websites, the correlation between the sadness and love emotion scores is the most notable. At this point, perhaps we might question the melancholy behind love (!?) and this might indicate a challenge in distinguishing these emotions or an interesting psychological linkage between them. Also, it's possible to say that our model doesn't distinguish these two emotions very well.

When we interpret this overall scenario, it's possible to say that for both websites, the emotion of joy is more differentiated from the other emotions and seems to be more distinctly expressed or identified in the reviews compared to other emotions across both websites.

❔Is there a pattern between higher-rated beer characteristics and reviews scored higher by positive emotions ?

Ecteristics

It aligns with our expectations that positive emotions, such as joy_score and love_score, demonstrate a positive correlation with beer characteristics, while most negative emotions exhibit negative correlations with these traits. In both the RB and BA datasets, we observe that love and joy scores show the highest positive correlations with these characteristics. This correlation seems reasonable as individuals tend to prefer beers with superior taste and aroma.

However, we also notice that the correlation between all emotion scores and appearance is not as straightforward. Considering the disparities between appearance and other beer characteristics, we hypothesise that this discrepancy might arise because when individuals comment on a beer, they often emphasise taste-related aspects rather than its visual appeal. 

❔How does ABV in beverages correlate with different emotions in user reviews?

E

In both the RB and BA datasets, the anger_score and fear_score exhibit the highest absolute correlations with beer ABV, and notably, these are the only high positive correlations observed. The intriguing similarity in the structures of both graphs is obvious, particularly the prominence of the anger_score. This observation could be attributed to the fact that higher ABV correlates with a higher alcohol percentage, and stronger alcohol content has the potential to evoke emotions, particularly making individuals more prone to feelings of anger. 

❔Is there a regional distribution characterizing users' emotions towards beer?

RateBeer_Love

BeerAdvocate_Suprise

The map shows a clear link between emotions and regions, with diverse color patterns in BA and RB. While some align, others like love and anger scores exhibit contrasting trends. Combining BA and RB data reduces biases, revealing deeper correlations between countries and emotional scores. Japan leads in anger_score in both datasets, while Malaysia tops fear_score and love_score in BA. Costa Rica stands out in the RB dataset with the highest sadness, fear, and love scores stemming from a single comment. These intriguing findings deepen our understanding of how countries relate to emotional scores on the map.

❔How emotion scores distribute across beer styles?

RateBeer

BeerAdvocate

Once we've calculated the average emotion scores for each type of beer, we observe how the scores are distributed across various styles of beer. 'Joy' emotion has the highest average score found in the RateBeer data sets, where averaged scores range from 0 to 0.25 for all emotions. Beer styles typically score lower on feelings like "sadness" and "fear." It's also possible to observe correlated emotions "love" and "anger" distributed between other emotions.

Beer Mood Mapping

It is pivotal to understand the distinction between moods and emotions, as they form the foundation of our analysis. 


Emotions are intense but fleeting, often a direct response to external stimuli. They are acute feelings like joy, anger, or surprise, and are typically short-lived. Emotions can be clearly identified and are usually associated with distinct physiological and behavioural changes.


Moods, on the other hand, are more subdued and enduring. They can last for hours, days, or even longer and often don't have a single identifiable cause. While emotions can be likened to the weather, striking distinctly and changing rapidly, moods are more comparable to the climate, a longer-term, pervasive state of mind. Moods are not as intense as emotions, but they can significantly influence a person's perception and interaction with the world.


In our project, we've analysed a vast array of beer reviews, extracting emotional scores using BERT-based semantic analysis. These scores, aggregated by beer and subsequently by beer style, have been crucial in constructing our 'Mood Map'. This map creates a unique correspondence between various moods and their associated high and low emotional states, enabling us to match beer styles with specific moods. This innovative approach offers a nuanced understanding of how different beers resonate with varying emotional landscapes, enhancing the beer selection experience based on mood congruence.

Here are the moods considered in our analysis:

1. Adventurous Sips: 

Imagine a beer that takes you on an expedition! Reviews often speak of 'adventurous' brews, capturing the essence of exploration and discovery. It's about beers that dare you to step out of your comfort zone, offering exotic and unexpected flavours.

4. Energetic Cheers: 

'Energetic' beers are all about vibrancy and liveliness. These reviews are filled with descriptions of bright, zesty, and invigorating beers that are like a burst of energy, ideal for lively parties or to kickstart a great night.

2. Cheerful Gulps: 

Then there are the 'cheerful' beers, often described in reviews as light, refreshing, and uplifting. These are the brews that bring smiles and laughter, perfect for a sunny day or a casual gathering with friends.


5. Nostalgic Reminiscence: 

Some beers take you on a trip down memory lane. These 'nostalgic' brews are often traditional, with a timeless taste that harks back to old memories and cherished moments.

3. Contemplative Reflections: 

On the flip side, some beers make us introspective and 'contemplative.' These are the reviews that talk about rich, complex beers that make you ponder life, often accompanied by deeper, more intense flavours.


6. Relaxed Sipping: 

Lastly, there are beers that embody relaxation. Reviews of 'relaxed' beers often mention smooth, mellow, and comforting flavours, ideal for unwinding after a long day.

For every mood considered, it's representation in the emotions space is defined in what we called, the ‘Mood Map’.

Results of the Mood Mapping

Mapping emotion scores to moods involves transforming complex semantic-similarity data into a more comprehensible and actionable format.


Through these steps, we have created a Mood Map that tries to resonate with the complexity and subtlety of human emotions and moods. Our analysis not only provides a unique perspective on beer preferences but also enhances the beer selection experience, allowing individuals to choose beers that align with their current mood or the ambience they wish to create.


The mood mapping results from both websites were meticulously combined, resulting in a consolidated list of beer styles categorised by mood, irrespective of the website they originated from. Out of a total of 169 beer styles analysed across these platforms, 74 were successfully mapped to specific moods. This translates to a mood mapping success rate of about 44%.


To effectively communicate the findings of our analysis, we have developed visual representations. These visuals succinctly illustrate the alignment of various beer styles with distinct moods. Focusing initially on the beer styles that have been mapped to particular moods, here is the distribution of these moods across the successfully categorised styles : 

The distribution of moods, as revealed by our analysis, is fairly balanced across five of the moods, with 'Adventurous' being an exception. This highlights the broad spectrum of emotional combinations and resulting moods elicited by various beer styles among consumers.


❔Which moods are most commonly associated with the best-selling beers?


Recall that earlier in our analysis, we identified the 20 most popular beer styles from each website. Based on the data, let's check out the relationship between these top-selling beer styles and their corresponding moods.

In our pie charts, each slice represents a mood and is weighted according to the popularity of the beer styles it encompasses. Here, popularity is gauged by the volume of reviews each style has received. A notable portion of beer styles is labelled as ‘Undetermined’ due to the limitations in our mood mapping process, which didn’t accommodate a mood categorization for every style. Nonetheless, it’s interesting to observe that the ‘Energetic’ mood stands out in both websites, reflecting a high preference for beer styles associated with this mood. In contrast, the 'Contemplative' mood has the smallest representation, indicating lesser popularity for beer styles that evoke a contemplative state. 

Don't forget to check out our table to select a beer style that suits your current mood and check to see if your favourite pub has it available 🍺!

Even though we're not beer experts, our list reveals a clustering of beer styles named with similar descriptive words. It appears that beer styles signified by the word "black" are associated with the mood of nostalgia, while beer styles denoted by the word "fruit" are matches with cheerful and adventurous moods. Interestingly, all stout beers match contemplative and nostalgic moods, showcasing how certain beer types can evoke specific emotional landscapes.

Conclusion

To bring our flavorful journey to a climax, we’ve brewed an interactive interface! This isn’t just any tool, it’s your personal mood-based beer concierge. Feeling adventurous or in need of relaxation? Just select your current mood and voilà: a curated list of beer styles, tailored to enhance your mood, will be at your fingertips. Dive into this immersive experience and discover beers that don’t just suit your taste but also match your mood, transforming the way you explore and enjoy beer. 


Imagine this: 

On a day fueled by a spirit of discovery, our interface suggests the 'Lambic - Fruit' for those in an 'Adventurous' mood. It's a choice that promises a taste escapade, with each sip offering a burst of unexpected fruity delight.😍 

Conversely, if serenity is what your heart seeks, a 'Relaxed' mood might lead you to the comforting embrace of an 'English Stout' or an 'Irish Dry Stout'. These selections are like a warm, soothing blanket for the soul, ideal for evenings spent in quiet contemplation or gentle unwinding.😊


Raise your glass to a new era of beer discovery, where each pour is more than just a drink—it's a narrative, a mood, a moment in time. Here's to adventures, relaxation, and everything in between. Cheers to a journey where every sip becomes a story!

Cheers!

References