Bell*, bellissim*

How is urban biodiversity perceived on Italian TikTok?

In contrast to previous studies, such as YouTube comments about North American mammals which observed reazioni negative which observed negative reactions to wild animals in urban environments, this analyses shows how Italian TikTok users tend to use positive adjectives, fostering empathy with the depicted animals.

Adjectives infact, offer a quick way to understand the visual, emotional, and characteristic traits that users most often associate with each animal.

One notable example is pigeons, whose reputation has shifted significantly—from being seen as innocent birds to being labeled "rats with wings".
On Italian TikTok pigeons receive primarily positive aesthetic descriptions, suggesting either a shift in perception or a platform-specific representation that differs from broader cultural narratives.

This phenomenon suggests not only a possible change in the collective perception of urban animals, but also how social media can influence and redefine established cultural narratives, opening new perspectives on the relationship between city and nature.

The predominance of positive descriptions, even for typically maligned species, may indicate that TikTok's "fun" culture is reshaping traditional perceptions of urban animals.

Which adjectives are used to describe animals on the Italian Tiktok?

To answer this question, we collected 2,030 terms from Italian TikTok videos about urban biodiversity and categorized them into five main groups.

Aesthetics

Quality

Size

Color

Observer

Other

Adjectives used at least 15 times in the dataset, ordered by frequency. The color represents the category.

Among these, adjectives related to positive aesthetic qualities are the most common.

Estetica

They are followed by the ones describing the size of animals, whether small or large. We noticed that Italian users tend to emphasize the dimensions of the animals.

Dimensioni

Clearly negative adjectives are less common overall.

Qualità

Colori

Osservatore

Altro

Aggettivi utilizzati almeno 15 volte all'interno del dataset, ordinati secondo la frequenza. Il colore rappresenta la categoria.

When looking among all animals, the most commonly used adjective is "bello" (beautiful).

However there are some exceptions, such as the seagull, for which the most frequently used adjective is “povero” (poor)...

...stemming from an Italian internet meme that began in 2021 featuring the popular song titled "Poor Seagull."

... and the bedbug, which is primarily described by its size (“piccolo,” small) followed by negative adjectives.

Likely due to the period of data collection, during which media and social networks amplified fears about the spread of a bedbug infestation originating in Paris throughout Europe, including Italy.

Some adjectives are common to all animals, while others are more distinctive for each animal.

To understand the most distinctive adjective for each animal, we reordered them using a TF-IDF (Term Frequency - Inverse Document Frequency), which isolates the most important words for each animal from those that are highly frequent across all animals.

The hawk for example is uniquely associated with the adjective “pilgrim”

Hawk

For squirrels, the color “grigio” (gray) is emphasized. This might be due to the fact that gray squirrels, although non-native to Italy, are more widespread and are altering the local ecosystem.

SQUIRREL

Or even, the crow is uniquely described as “intelligente” (smart).

CROW

Notes on data collection and analysis

The data was collected using the Zeeschuimer, browser extension, which enables systematic data collection while browsing social media. To minimize the effects of TikTok's algorithmic personalization, data was collected using a "clean search browser" and a new TikTok account created specifically for this study.

The search queries were generated by combining:
→ A list of animals (both singular and plural)
→ A list of Italian cities (regional capitals)

From an initial dataset of 41,129 videos, only those meeting the following criteria were selected:
‍→ Videos clearly depicting urban animals
→ Videos set in an Italian urban context
→ Videos with verifiable geolocation or an identifiable Italian urban setting

The final dataset consists of 2,648 videos, including:
‍→ 1.600 con città chiaramente identificata
→ 1,048 set in an unspecified Italian urban context

The analysis of adjectives was carried out using the spaCy natural language processing library.

The adjectives were unified according to the following criteria:
→ Unification of plurals and singulars
→ Unification of masculine and feminine forms
→ Retention of basic forms of adjectives