<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Method of preprocessing information for preparing a description of art objects using artificial intelligence⋆</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Voichur</string-name>
          <email>o.voichur@gmail.com</email>
          <email>voichury@khmnu.edu.ua</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Hovorushchenko</string-name>
          <xref ref-type="aff" rid="aff1">1</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Artem</string-name>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Anatoliy Nester</string-name>
          <email>nesteranatol111@gmail.com</email>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>Khmelnytskyi National University</institution>
          ,
          <addr-line>Institutska str., 11, Khmelnytskyi, 29016</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
        <aff id="aff1">
          <label>1</label>
          <institution>National Pirogov Memorial Medical University</institution>
          ,
          <addr-line>Pirogova str., 56, Vinnytsya, 21018</addr-line>
          ,
          <country country="UA">Ukraine</country>
        </aff>
        <aff id="aff2">
          <label>2</label>
          <institution>Tallinna Tehhnikaülikool</institution>
          ,
          <addr-line>Ehitajate tee 5, Tallinn, 12616</addr-line>
          ,
          <country country="EE">Estonia</country>
        </aff>
      </contrib-group>
      <abstract>
        <p>The use of generative artificial intelligence (AI) in the preparation of artwork descriptions allows automating the process of analyzing and documenting works, improving the accuracy and quality of information. Creating textual descriptions for visually impaired people is a combination of the art of words and inclusive technologies, as it is important not only to convey the meaning of the painting but also to help a person “see” it through emotions, analogies, and sensory sensations. The introduction of this technology will help improve the perception of art, its accessibility to a wide audience, and the quality of educational materials, and proper data preprocessing is a key factor in obtaining relevant and accurate descriptions, which contributes to the development of digital archives, museum collections, and educational platforms. The main goal of the developed method of preprocessing information for preparing art object descriptions using artificial intelligence is to adapt the text to the individual characteristics of user perception, including their emotional, cognitive, and cultural preferences and gender identity. The language models/prompts for the automatic generation of personalized, emotionally colored textual descriptions of art objects obtained as a result of the developed method, as well as the results of their implementation, allow to supplement information technology for ensuring accessibility to art objects for the visually impaired persons with the following functions: generation of personalized, adapted textual descriptions of art objects, taking into account user preferences and gender; generation of detailed audio descriptions of paintings using AI, taking into account the context and emotional component of the work, as well as the ability to adapt the description style to user preferences (for example, a choice between a detailed and concise description); interactive interaction of users with the system through voice commands or textual queries; integration with existing inclusive technologies for visually impaired people.</p>
      </abstract>
      <kwd-group>
        <kwd>eol&gt;Information preprocessing</kwd>
        <kwd>description of art objects</kwd>
        <kwd>artificial intelligence (AI)</kwd>
        <kwd>generative artificial intelligence (GAI)</kwd>
        <kwd>psychophysiology of attention</kwd>
        <kwd>focus of attention</kwd>
        <kwd>1</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
      <p>
        In 2015, there were an estimated 253 million people with visual impairments worldwide, of whom 36
million were totally blind and 217 million had moderate to severe visual impairment [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ]. By 2020,
this number had increased to 295 million people, including 43.3 million blind and 251.7 million people
with severe visual impairment [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]. It is projected that by 2050, the total number of blind or
moderately and severely visually impaired people could reach 703 million [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ].
      </p>
      <p>According to official statistics, there are approximately 70,000 blind people in Ukraine today, but
unofficial estimates suggest that this figure may be three times higher.</p>
      <p>
        The ratification of the UN Convention on the Rights of Persons with Disabilities obliges Ukraine
to create favorable conditions and opportunities for the development of people with disabilities as
part of the country's sustainable development [
        <xref ref-type="bibr" rid="ref3">3</xref>
        ]. In particular, according to Article 30 of the
Convention, the state must ensure accessibility of television, cinema, theatrical performances and
other cultural events for persons with disabilities; the possibility of visiting cinemas, museums,
libraries and other cultural institutions; conditions for creative self-realization of persons with
disabilities; access to cultural heritage for persons with various forms of disability; adaptation of
cultural works for visually impaired persons without infringing copyright.
      </p>
      <p>
        Today, in Ukraine, as well as around the world, painting and visual arts in general remain almost
inaccessible to people with visual impairments. At the same time, two-thirds (66.9%) of such people
in Ukraine consider art and participation in cultural life important to them: 42% completely agree
with this, and another 24.9% rather agree [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ]. In addition, almost 62% of people with visual
impairments are convinced that the state should ensure equal rights for people with disabilities in
the field of culture, which is in line with the general sentiments of Ukrainian society [
        <xref ref-type="bibr" rid="ref4 ref5">4, 5</xref>
        ].
      </p>
      <p>
        Modern medical information technologies play an important role in the lives of visually impaired
people [
        <xref ref-type="bibr" rid="ref6 ref7">6, 7</xref>
        ]. Medical information technologies greatly facilitate the lives of visually impaired people,
enabling them to integrate into society, access information and actively participate in cultural,
professional and personal life [
        <xref ref-type="bibr" rid="ref8 ref9">8, 9</xref>
        ]. Thanks to innovative technologies, visually impaired people can
significantly improve their quality of life, increase their level of independence, and take an active
part in public life. This helps to reduce barriers and ensure equal opportunities for all people,
regardless of their physical disabilities.
      </p>
      <p>Artificial intelligence can be used to adapt media resources to the needs of visually impaired
people, for example, by creating automated descriptions for paintings, accurate audio or tactile
descriptions for accessing cultural objects and other elements of the environment.</p>
      <p>
        Thus, it is obvious that creating an accessible artistic environment for visually impaired people is
an important task today. This can be realized by improving the previously designed information
technology [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ], which will not only transform 2D images into 3D models, but also generate
descriptions using artificial intelligence, which will be converted into Braille using specialized
software and into audio recordings.
      </p>
    </sec>
    <sec id="sec-2">
      <title>2. Psychophysiology of attention. Focus of attention depending on gender.</title>
      <p>
        The physiological basis of attention is the processes of excitation and inhibition and the peculiarities
of their mobility and interaction in the cerebral cortex [
        <xref ref-type="bibr" rid="ref10">10</xref>
        ].
      </p>
      <p>
        The directionality of higher nervous and mental activity is always associated with the excitation
of some cortical areas and the inhibition of others according to the law of induction [
        <xref ref-type="bibr" rid="ref11 ref12">11, 12</xref>
        ].
      </p>
      <p>Among the excited cortical areas, the one that is most important at that moment stands out and
begins to dominate all others. This ensures the selectivity of our activity and control over its course,
so we can keep our attention on an object for a long time [13].</p>
      <p>Any selective activity of the brain is associated with a certain level of its activity, which in turn
is set by a special brain apparatus, including the reticular formation and the frontal lobes of the brain
[14, 15].</p>
      <p>Brain activation can be associated with physiological needs or environmental stimuli [16].</p>
      <p>Stimuli can affect brain activity in two ways: through the reticular formation; through specific
sensory areas of the cortex and frontal lobes.</p>
      <p>This mechanism of brain activation is the basis of involuntary attention (attention caused by
external causes – certain features of objects that affect a person at that moment) [17, 18].</p>
      <p>Voluntary attention (active; a controlled and conscious process in which the subject actively
chooses an object that is meaningful to him or her) is associated with the activity of the frontal lobes
of the cerebral hemispheres and the formation of a dominant in a certain center of the brain
(dominance of the focus of excitation) [19].</p>
      <p>Both attention mechanisms also include the limbic system, which provides vegetative and
emotional support for mental activity.</p>
      <p>A person's focus of attention may differ by gender due to biological, cognitive, and social factors.
Men and women can perceive and interpret art differently, paying attention to different aspects of
works. This is due to the influence of social, cultural and psychological factors that shape their
aesthetic preferences and perceptions. Women are characterized by distributed attention (they are
able to concentrate on several tasks at the same time – multitasking); they notice small details,
nonverbal cues and emotions of the interlocutor better (context); focus on interpersonal interactions,
relationships and non-verbal communication (social orientation); notice changes in the environment
faster, especially in social or emotional aspects. Men have a stronger ability to tunnel vision
(concentration on a single task); are better at spatial orientation, noticing global structures and
directions (spatial orientation); tend to pay attention to logical connections and mechanical details
(analytical approach); may be less emotional about facial expressions and gestures (less sensitive to
social cues). However, these features are generalizations, and the focus of attention is also influenced
by individual characteristics, experience, and training [20-22].</p>
      <p>Other key factors also influence the characteristics and focus of attention [23-25]:
•
•
•
•
•</p>
      <p>Hormones – estrogen and testosterone levels can affect cognitive processes. For example,
estrogen improves verbal memory and the ability to multitask, while testosterone promotes
focus on a single task.</p>
      <p>Evolutionary features – some researchers suggest that historically, women have developed
distributed attention skills due to the need to care for children and manage the household,
while men have had an advantage in narrow focus for hunting or defense.</p>
      <p>Social and cultural factors – upbringing and societal expectations influence how men and
women develop their attention. For example, girls are more likely to be taught to be attentive
to the emotions of others, while boys are more likely to be taught to be attentive to technical
details.</p>
      <p>Neuropsychology – MRI brain scans show that women have more active areas related to
verbal communication, and men have more active areas related to spatial thinking. This may
explain the difference in attention span.</p>
      <p>Personality traits – regardless of gender, people differ in their attention types (concentrated,
distributed, selective, etc.) due to genetics, experience, and professional activities.</p>
      <p>Given these key factors, men and women can perceive paintings differently due to the
peculiarities of attention, emotional perception, and cognitive processes [26, 27].</p>
      <p>Women pay attention to [28-30]:
•
•
•</p>
      <p>Emotional mood and atmosphere – women are more likely to analyze the feelings evoked by
a painting, pay attention to the color scheme and light play, facial expressions, character
interaction, color and texture nuances that evoke feelings of compassion, care or tenderness;
they pay more attention to the softness and depth of emotions conveyed through the
composition.</p>
      <p>Minor details, context and symbolism – they are interested in minor elements that may carry
hidden meaning or reinforce the overall message of the picture; they are more inclined to
interpret symbolism related to personal or social aspects of the picture; they may focus on
those details that express subtle social or psychological messages, relationships between
characters, as well as aspects related to human emotions and the inner world.</p>
      <p>Images of people and their emotions – women are more likely to read facial expressions,
gestures, and postures of characters; are more interested in studying the interaction of
•
•
•
characters in a picture, especially if it concerns emotional connections or social relationships;
pay attention to details related to social roles, facial expressions, and bodily movements that
reflect relationships between people.</p>
      <p>Harmony and aesthetics – they evaluate the smoothness of lines, softness of transitions
between shades, and overall compositional balance.</p>
      <p>Color and texture – they often respond more emotionally to the color palette, especially to
shades that can evoke feelings of warmth, calmness, or even melancholy; may pay attention
to textures and smooth color transitions that create a sense of depth and tenderness.
Personal preferences and social stereotypes – may be more interested in details that depict
human interaction and emotional depth, as they traditionally place more emphasis on human
relationships.</p>
      <p>Men pay attention to [28-30]:
•
•
•
•
•</p>
      <p>Composition and structure – men are more focused on the construction of the scene, the
logic of the arrangement of elements, perspectives; more likely to focus on aspects that reflect
strength, dynamics or energy, such as color contrasts, compositional elements that create a
sense of tension or movement; pay more attention to structural details and elements that
convey activity or dynamism.</p>
      <p>Dynamics and movement – they are more likely to notice tension in poses, character
interactions, and overall plot development; focus on more “active” aspects such as lines,
shapes, and objects that create images of strength, struggle, or action; may pay attention to
proportions and structural elements that give a painting visual strength and impact.
Contrasts and technique – men may pay attention to the play of light and shadow, expressive
brushstrokes, clean lines, and texture; may focus on color contrasts, bright and saturated hues
that draw attention and create more intense visual effects; pay attention to sharpness and
clarity of lines that enhance the sense of movement or force.</p>
      <p>Plot and logic of events – it is important for them to understand what exactly is happening
in the picture, how characters and objects are connected to each other; they may pay more
attention to the overall context and compositional elements, in particular to the aspect of
historical or cultural significance, where attention is focused on the big picture, dynamics, or
images of struggle or victory.</p>
      <p>Personal preferences and social stereotypes – may be more focused on aspects related to
activity, movement, large spaces, strength, and interaction with objects.</p>
      <p>Of course, these are not hard and fast rules, but generalizations that may differ depending on the
individual and cultural background. The true perception of a painting depends on the individual
experience and the socio-cultural context in which the viewer is located. Gender can influence which
aspects of a painting are particularly salient through emotional, symbolic, or other socially
determined factors. This approach allows for a better understanding of how different audiences may
perceive art in different ways and helps to create more adapted and inclusive descriptions for
different groups of viewers. It is these differences that are taken into account, for example, in
advertising, design, and the creation of visual content for different audiences. These differences
should also be taken into account when creating a description of an art object (painting), as the
description of a painting should change depending on whether it is aimed at men or women, as men
and women often pay attention to different aspects of a painting.</p>
      <p>An example of a painting description for women: “The painting radiates tenderness and depth;
every stroke conveys the elegance of details. Soft colors create an atmosphere of peace and harmony,
and smooth lines add emotional warmth to the composition. An important role is played by the play
of light and shadow, which emphasizes the subtlest shades of feelings. The heroine (or the central
image) is filled with inner peace, her gaze is mysterious, and her posture is natural and relaxed. The
painting seems to invite personal reflection, awakening memories or dreams”. An example of a
description of the painting for men: “The painting impresses with its dynamics and compositional
balance. Contrasting colors and sharp lines create the effect of depth, and the accuracy of the details
adds to the realism. The central image or the main figure stands out due to the light and shadow
accents, which makes the scene lively and full of character. There are no random elements here
every detail is subject to the general idea, creating a strong visual effect. The painting arouses interest
in the details and technique of execution, encouraging a logical analysis of its content”.</p>
      <p>Both descriptions reflect the same painting, but emphasize different details according to the
perception of men and women. Although these descriptions are generalized, they demonstrate that
men and women's attention can focus on different aspects: women are more likely to notice the
emotional component, atmosphere, and subtle details, while men are more focused on structure,
composition, and technique. Thus, research shows that men and women may have different
preferences in terms of presentation style, word choice, and aspects of art that interest them, which
requires personalization of content, including by gender.</p>
    </sec>
    <sec id="sec-3">
      <title>3. Method of preprocessing information for preparing a description of art objects using artificial intelligence</title>
      <p>Modern artificial intelligence technologies open up new opportunities for analyzing and describing
artworks. Generative artificial intelligence (GAI) allows automating the process of generating textual
descriptions of paintings, sculptures, and other works of art. This helps to expand the accessibility
of cultural heritage for the visually impaired. However, creating textual descriptions of paintings for
visually impaired people requires a special approach. Such a description should be not just
informative, but also figurative, emotionally colored, detailed, and understandable, so that the
listener can “feel” the painting through words. One of the important aspects of developing such
descriptions is personalizing the content in accordance with the peculiarities of information
perception by different audiences, in particular, taking into account gender differences. Therefore,
taking into account gender specifics can improve the user experience and increase engagement in
the study of art objects. Therefore, the effective use of GAI requires high-quality preprocessing of
incoming information.</p>
      <p>The main stages of the method of preprocessing information for preparing a description of art
objects using artificial intelligence:</p>
      <p>Input: digital or 3D images of art objects.
1. Data collection and preprocessing – collection of digital or 3D images of art objects,
accompanying metadata (author, year of creation, dimensions, materials, style, etc.) and
available descriptions from various sources (museum catalogs, literary descriptions, scientific
articles); processing of text, graphic and metadata; data cleaning and normalization
(elimination of noise, duplicates, inaccuracies; conversion of unstructured data into a
structured format); data formatting (bringing images to a single format).
2. Image segmentation – division of an image into areas that differ in certain characteristics;
these areas correspond to real objects or their parts, and their boundaries coincide with the
contours of the objects. The main purpose of segmentation is to simplify or change the image
representation in such a way as to facilitate its further analysis. This stage includes automatic
or semi-automatic separation of the picture into different components or objects that may be
important for further description: highlighting the main parts of the picture (characters,
objects, background); identification of various aspects of the composition (for example, color
contrasts, textures, elements in the foreground and background).
3. Image analysis – computer vision and image analysis methods can be used to obtain
comprehensive information about the painting: object recognition (identifying the main
elements in the painting), color scheme determination (classification of primary colors, their
combination and shades used), structural analysis (determining the compositional features of
the painting, its perspective, chiaroscuro and texture), contextual analysis (identification of
style, historical period, context of creation).
4. Identification of key characteristics of the art object – analysis of shape and composition
(computer vision methods, for example, OpenCV), color analysis (clustering by color,
contrast and lighting analysis), texture analysis (identifying features of brush strokes, surface
material), object-oriented analysis (recognition of characters, objects, background).
5. Semantic analysis and classification – defining categories (genre, style, technique) using
neural network classification models (ResNet, EfficientNet, VGG-16).
6. Determination of the target audience for further adaptation for different categories of users
(for example, taking into account their gender) – before generating a description, it is
important to determine what it is intended for and for whom, as the style and level of detail
may differ: scientific research – detailed technical description, references to sources, art
historical analysis; museum catalogs – structured description with an emphasis on historical
and cultural context; educational materials – simplified but informative presentation of
material for students and teachers; popular resources – emotionally colored descriptions
adapted for the general public (this style is especially important when describing art objects
for the visually impaired); women (more likely to notice the emotional component,
atmosphere, and subtle details, so it is important to focus on emotionality, symbolism,
connection to everyday life, social and cultural aspects) or men (more focused on structure,
composition, and technique, so it is important to focus on historical facts, technical
characteristics, analysis of artistic techniques and styles), etc.
7. Choosing the style and tone of the text – generative AI can adapt the style of the text
depending on the task: formal style – an academic approach using specialized terminology;
descriptive-artistic style – an emotionally expressive text that conveys the atmosphere of the
work, the impressions that the work evokes in the viewer, what feelings it evokes in the
viewer (style is especially important when building a description of art objects for people
with visual impairments); commercial style – an emphasis on the uniqueness and value of
the work for auctions, sales.
8. Generating the structure of the description – determining the title, author, year of creation;
building a general compositional scheme with a description of key elements.
9. Forming the substantive part of the description – determining the main plot of the work,
detailing the elements of the painting, including a description of the main objects and their
mutual arrangement, revealing the meaning and emotional component of the work of art,
analyzing the color palette and painting style.
10. Ensuring adaptation for people with visual impairments – using accessible descriptions of
paintings and adapting them to the standards of inclusive description, formulating clear and
detailed text explanations for audio description, using terminology that contributes to the
perception of the texture, shape, space and emotional coloring of the depicted. The main
principles of forming a description for people with visual impairments: clarity and structure
(the description should be logical and understandable, divided into blocks), imagery and
detail (using metaphors, comparisons, descriptions of texture, depth, colors and light and
shade), using sensory analogies (comparison with tactile or sound sensations), lack of
excessive terminology (if art terms are used, they should be explained), preserving the
author's intention (transmitting the emotional impact of the painting).
11. Developing a language model / prompt for automatic generation of text descriptions –
configuring generative AI to create natural and meaningful descriptions; training the model
on a large volume of artistic descriptions that take into account style, historical context,
artistic means; using quality control methods to check the accuracy and correspondence of
the descriptions to the original works.</p>
      <p>Output: language model / prompt for automatic generation of personalized text description
of art objects.</p>
      <p>The use of generative artificial intelligence (AI) in the preparation of artwork descriptions allows
automating the process of analyzing and documenting works, improving the accuracy and quality of
information. Creating textual descriptions for visually impaired people is a combination of the art of
words and inclusive technologies, as it is important not only to convey the meaning of the painting
but also to help a person “see” it through emotions, analogies, and sensory sensations. The
introduction of this technology will help improve the perception of art, its accessibility to a wide
audience, and the quality of educational materials, and proper data preprocessing is a key factor in
obtaining relevant and accurate descriptions, which contributes to the development of digital
archives, museum collections, and educational platforms. The main goal of the developed method of
preprocessing information for preparing art object descriptions using artificial intelligence is to
adapt the text to the individual characteristics of user perception, including their emotional,
cognitive, and cultural preferences and gender identity.</p>
    </sec>
    <sec id="sec-4">
      <title>4. Results</title>
      <p>As a result of the research, two gender-oriented prompts were created to create personalized
descriptions of Vincent van Gogh's painting “Starry Night” for people with visual impairments.</p>
      <p>Thus, a gender-oriented prompt for creating a personalized emotionally colored description of
Vincent van Gogh's painting “Starry Night” for women with visual impairments looks like this:
“Create an emotional and sensual description of Vincent van Gogh's painting “Starry Night,”
focusing on its symbolism, emotional color scheme, and the impact of the images on the viewer's
mood. Create a detailed verbal description of the painting for people with visual impairments. Use
emotional and figurative language to convey the atmosphere, color palette, and plot of the work.
Avoid complex terms, make the text accessible and understandable”. The result of executing such a
prompt in ChatGPT is presented in Fig. 1.</p>
      <p>A gender-specific prompt for creating a personalized description of Vincent van Gogh's Starry
Night through the prism of history, technique, and artistic style for men with visual impairments
looks like this: “Briefly describe Vincent van Gogh's Starry Night in terms of history, technique, and
artistic style. Focus on the details of composition, perspective, and context. Create a detailed verbal
description of the painting for people with visual impairments”. The result of executing such a
prompt in ChatGPT is presented in Fig. 2.</p>
      <p>Taking into account the research conducted in Chapter 2 on how gender affects the focus of
attention when studying a painting, a gender-based segmentation of Leonardo da Vinci's Mona Lisa
was also conducted and two gender-based personalized descriptions of the painting were created for
people with visual impairments. Description of the painting "Mona Lisa" by Leonardo da Vinci for
women with visual impairments taking into account gender-based segmentation of the image (Fig.
3) and with an emphasis on emotions, textures and soft details:
•
•
•
•</p>
      <p>Upper part (face and hair): The focus is on the woman's face. It is oval with soft, smooth lines.
The skin is light, smooth in appearance. The eyes are dark, their gaze is gentle and calm, creating
the feeling that she is looking right at you. The lips are slightly raised in a mysterious smile, which
arouses curiosity. The hair is dark, smooth, falls in soft waves to the shoulders, adding femininity.
Middle part (hands and clothes): Her hands are folded one on top of the other on her lap –
this pose creates an impression of calm and restraint. The fingers are long and thin. The
clothing is dark in color, the fabric looks dense, but with soft folds. Sheer fabric is visible on
the chest, which adds lightness to the image
Lower part (background and surroundings): The background is a landscape with soft shapes:
calm lines of roads, a winding river and distant mountains. The colors of the background are
muted, which helps to highlight the figure of the woman in the foreground. The nature
behind her seems to complement the calm nature of the portrait.</p>
      <p>Focus for women: gentle gaze, softness of facial lines, texture of hair and a sense of calm in
the position of the hands.</p>
      <p>Description of the painting "Mona Lisa" by Leonardo da Vinci for men with visual impairments,
taking into account the gender-oriented segmentation of the image (Fig. 4) and with an emphasis on
structure, spatial arrangement and contrast:
•</p>
      <p>Upper part (head and shoulders): The woman's face is located in the upper central part of the
canvas. Its shape is oval with clear but soft contours. The eyes are located on a horizontal
axis and look straight ahead, which creates a focus for the viewer. Above the eyebrows is a
smooth curve of the forehead. The lips are compressed, forming a barely noticeable smile.</p>
      <p>The hair is dark, symmetrically framing the face.
•
•
•</p>
      <p>Middle part (torso and arms): The body is turned at a slight angle to the left, but the head is
pointed straight ahead. This creates a dynamic composition. The arms are folded in front,
forming a horizontal line that balances the vertical lines of the body. The clothing is dark
with clearly defined folds that add texture.</p>
      <p>Lower part (background and perspective): The background starts at about shoulder level and
extends into the background. Behind is a series of winding lines representing roads and a
river that lead the eye into the distance. Mountains on the horizon create depth and
perspective. The contrast between the dark figure and the lighter background emphasizes the
main subject.</p>
      <p>Focus for men: the structure of the figure, the spatial arrangement of elements, the clarity of
lines, and the contrast between the foreground and background.</p>
      <p>
        The language models/prompts for the automatic generation of personalized, emotionally colored
textual descriptions of art objects obtained as a result of the developed method, as well as the results
of their implementation, allow to supplement information technology for ensuring accessibility to
art objects for the visually impaired persons, early designed in [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ], with the following functions:
      </p>
      <p>Generation of personalized, adapted textual descriptions of art objects, taking into account
user preferences and gender.</p>
      <p>Generation of detailed audio descriptions of paintings using AI, taking into account the
context and emotional component of the work, as well as the ability to adapt the description
style to user preferences (for example, a choice between a detailed and concise description).
Interactive interaction of users with the system through voice commands or textual queries.</p>
      <p>Integration with existing inclusive technologies for visually impaired people.</p>
    </sec>
    <sec id="sec-5">
      <title>5. Conclusions</title>
      <p>
        An important task today is to create an accessible art environment for people with visual
impairments. This can be implemented by improving the previously designed information
technology for ensuring accessibility to art objects for the visually impaired persons [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ], which will
transform 2D images into 3D models and use artificial intelligence to generate their description,
which will be converted into Braille using specialized software and into audio recordings.
      </p>
      <p>Gender can affect which aspects of a painting attract special attention, due to emotional, symbolic
or other socially determined factors. Men and women may focus on different aspects: women are
more likely to notice the emotional component, atmosphere and fine details, while men focus more
on structure, composition and execution technique. Thus, research suggests that men and women
may have different preferences for presentation style, choice of words and aspects of art that interest
them, which requires personalization of content, including depending on gender. This approach
allows for a better understanding of how different audiences may perceive art differently, and helps
create more tailored and inclusive descriptions for different groups of viewers.</p>
      <p>The use of generative artificial intelligence (AI) in the preparation of artwork descriptions allows
automating the process of analyzing and documenting works, improving the accuracy and quality of
information. Creating textual descriptions for visually impaired people is a combination of the art of
words and inclusive technologies, as it is important not only to convey the meaning of the painting
but also to help a person “see” it through emotions, analogies, and sensory sensations. The
introduction of this technology will help improve the perception of art, its accessibility to a wide
audience, and the quality of educational materials, and proper data preprocessing is a key factor in
obtaining relevant and accurate descriptions, which contributes to the development of digital
archives, museum collections, and educational platforms. The main goal of the developed method of
preprocessing information for preparing art object descriptions using artificial intelligence is to
adapt the text to the individual characteristics of user perception, including their emotional,
cognitive, and cultural preferences and gender identity.</p>
      <p>
        The language models/prompts for the automatic generation of personalized, emotionally colored
textual descriptions of art objects obtained as a result of the developed method, as well as the results
of their implementation, allow to supplement information technology for ensuring accessibility to
art objects for the visually impaired persons, early designed in [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]m with the following functions:
generation of personalized, adapted textual descriptions of art objects, taking into account user
preferences and gender; generation of detailed audio descriptions of paintings using AI, taking into
account the context and emotional component of the work, as well as the ability to adapt the
description style to user preferences (for example, a choice between a detailed and concise
description); interactive interaction of users with the system through voice commands or textual
queries; integration with existing inclusive technologies for visually impaired people.
      </p>
    </sec>
    <sec id="sec-6">
      <title>Acknowledgements</title>
      <p>The authors would like to thank the EACEA and the ERASMUS+ SMART-PL project for the idea,
inspiration and equipment that made this work possible.</p>
    </sec>
    <sec id="sec-7">
      <title>Declaration on Generative AI</title>
      <p>During the preparation of this work, the authors used Grammarly in order to: grammar and spelling
check; DeepL Translate in order to: some phrases translation into English; ChatGPT in order to:
conduct experiments as a prompt-based tool for creating automated descriptions of art objects. After
using these tools/services, the authors reviewed and edited the content as needed and take full
responsibility for the publication’s content.
[13] C. A. Miller, R. Hübner, The Relations of Empathy and Gender to Aesthetic Response and
Aesthetic Inference of Visual Artworks, Empirical Studies of the Arts 41 1 (2022) 188-215. doi:
10.1177/02762374221095701.
[14] A. Miscenà, V. Frentzen, J. Arato, Z. Dare, H. Leder, R. Rosenberg, No such thing as the female
eye: ditching gender-binary categories in art perception, Fem. Media Stud. (2024) 1–25.
doi:10.1080/14680777.2024.2361043.
[15] A. Larrain, A. Haye, The dialogical and political nature of emotions: A reading of Vygotsky’s</p>
      <p>The Psychology of Art, Theory &amp; Psychol. 30.6 (2020) 800–812. doi:10.1177/0959354320955235.
[16] F. González Rey, Vygotsky’s Concept of Perezhivanie in The Psychology of Art and at the Final
Moment of His Work: Advancing His Legacy, Mind, Cult., Act. 23.4 (2016) 305–314.
doi:10.1080/10749039.2016.1186196.
[17] G. Mather, Psychology of Art, Taylor &amp; Francis Group, 2020. 144 p.
[18] F. L. GONZÁLEZ REY, Vygotsky’s “The Psychology of Art”: A foundational and still unexplored
text, Estud. Psicol. (Camp.) 35.4 (2018) 339–350. doi:10.1590/1982-02752018000400002.
[19] M. Martín, M. D. Valiña, Heuristics, Biases and the Psychology of Reasoning: State of the Art,</p>
      <p>Psychology 14.02 (2023) 264–294. doi:10.4236/psych.2023.142016.
[20] J. M. Hansen, T. Roald, Aesthetic Empathy: An Investigation in Phenomenological Psychology
of Visual Art Experiences, Journal of Phenomenological Psychology (2022).
[21] R. M. Rodriguez-Boerwinkle, M. J. Boerwinkle, P. J. Silvia, The Open Gallery for Arts Research
(OGAR): An open-source tool for studying the psychology of virtual art museum visits, Behav.</p>
      <p>Res. Methods (2022). doi:10.3758/s13428-022-01857-w.
[22] K. Oatley, M. Djikic, Psychology of Narrative Art, Rev. Gen. Psychol. 22.2 (2018) 161–168.</p>
      <p>doi:10.1037/gpr0000113.
[23] G. Minissale, From relational aesthetics to relational knowledge, in: The Psychology of
Contemporary Art, Cambridge University Press, Cambridge, с. 332–346.
doi:10.1017/cbo9781139094313.044.
[24] G. Razali, F. S. Mutma, N. P. Adriana, D. Angelina, Z. S. Saldi, Creative Communication as a
Catalyst for Change: Shaping Urban Development Through Psychology and Art, Communica
1.1 (2023) 31–42. doi:10.61978/communica.v1i1.174.
[25] E. Li, Research on Visual Expression of Color Collocation in Art Education Based on Art</p>
      <p>Psychology, Int. J. Educ. Humanit. 3.3 (2022) 27–31. doi:10.54097/ijeh.v3i3.1005.
[26] M. Skov, M. Nadal, A Farewell to Art: Aesthetics as a Topic in Psychology and Neuroscience,</p>
      <p>Perspect. Psychol. Sci. 15.3 (2020) 630–642. doi:10.1177/1745691619897963.
[27] M. Orr, Towards a feminist revisionism of an aesthetics of mastery, in: Reading, writing and the
influence of Harold Bloom, Manchester University Press, 2024.
doi:10.7765/9781526186027.00014.
[28] N. A. Michna, Feminist aesthetics: then and now – reflections on thirty-five years of inquiry in
the US tradition, Fem. Theory (2024). doi:10.1177/14647001241284969.
[29] S. Cefai, Feminist Aesthetics of Resistance, in: The Routledge Companion to Gender and Affect,</p>
      <p>Routledge, London, 2022, с. 227–236. doi:10.4324/9781003045007-25.
[30] D. Harris, Indigenous Feminist Aesthetic Work as Cultural Revitalization: Facilitating
Uy’Skwuluwun, in: Feminism, Adult Education and Creative Possibility, Bloomsbury Academic,
2022. doi:10.5040/9781350231078.ch-11.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <given-names>P.</given-names>
            <surname>Ackland</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Resnikoff</surname>
          </string-name>
          ,
          <string-name>
            <given-names>R.</given-names>
            <surname>Bourne</surname>
          </string-name>
          .
          <article-title>World blindness and visual impairment: Despite many successes, the problem is growing</article-title>
          .
          <source>Community Eye Health Journal</source>
          (
          <year>2018</year>
          )
          <fpage>71</fpage>
          -
          <lpage>73</lpage>
          . PMID:
          <volume>29483748</volume>
          .
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <article-title>GBD 2019 Blindness and Vision Impairment Collaborators; Vision Loss Expert Group of the Global Burden of Disease Study. Trends in prevalence of blindness and distance and near vision impairment over 30 years: an analysis for the Global Burden of Disease Study</article-title>
          .
          <source>Lancet Glob Health</source>
          (
          <year>2021</year>
          )
          <fpage>e130</fpage>
          -
          <lpage>e143</lpage>
          . doi:
          <volume>10</volume>
          .1016/
          <fpage>S2214</fpage>
          -109X(
          <issue>20</issue>
          )
          <fpage>30425</fpage>
          -
          <lpage>3</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>P.</given-names>
            <surname>Hryhoruk</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S.</given-names>
            <surname>Grygoruk</surname>
          </string-name>
          ,
          <string-name>
            <given-names>N.</given-names>
            <surname>Khrushch</surname>
          </string-name>
          ,
          <string-name>
            <given-names>T.</given-names>
            <surname>Hovorushchenko</surname>
          </string-name>
          .
          <article-title>Using non-metric multidimensional scaling for assessment of regions' economy in the context of their sustainable development</article-title>
          .
          <source>CEUR-WS</source>
          <volume>2713</volume>
          (
          <year>2020</year>
          )
          <fpage>315</fpage>
          -
          <lpage>333</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          <article-title>[4] ART FOR ALL: THE SITUATION WITH THE OBSERVANCE OF CULTURAL RIGHTS OF PEOPLE WITH DISABILITIES IN UKRAINE. Analytical report based on the results of the allUkrainian survey "Opinions and Views of the Population of Ukraine" (Omnibus) in September 2021</article-title>
          . URL: https://ffr.org.ua/wp-content/uploads/2022/10/Mystetstvo-dlya-vsih_
          <article-title>-sytuatsiya-zdotrymannyam-kulturnyh-prav-lyudej-z-invalidnistyu-v-Ukrayini.pdf_</article-title>
          .pdf. [in Ukrainian]
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          [5]
          <string-name>
            <given-names>T.</given-names>
            <surname>Hovorushchenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>O.</given-names>
            <surname>Voichur</surname>
          </string-name>
          ,
          <string-name>
            <given-names>O.</given-names>
            <surname>Hovorushchenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Boyarchuk</surname>
          </string-name>
          ,
          <string-name>
            <surname>I. Zasornova</surname>
          </string-name>
          ,
          <article-title>The Concept of Information Technology for Ensuring Accessibility to Art Objects for the Visually Impaired Persons</article-title>
          ,
          <source>CEUR-WS</source>
          <volume>3675</volume>
          (
          <year>2024</year>
          )
          <fpage>208</fpage>
          -
          <lpage>222</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <given-names>T.</given-names>
            <surname>Hovorushchenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Moskalenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Osyadlyi</surname>
          </string-name>
          ,
          <article-title>Methods of medical data management based on blockchain technologies</article-title>
          ,
          <source>J. Reliab. Intell. Environ</source>
          . (
          <year>2022</year>
          ).
          <source>doi:10.1007/s40860-022-00178-1.</source>
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <given-names>T.</given-names>
            <surname>Hovorushchenko</surname>
          </string-name>
          , Ye. Hnatchuk,
          <string-name>
            <given-names>A.</given-names>
            <surname>Herts</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Moskalenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>V.</given-names>
            <surname>Osyadlyi</surname>
          </string-name>
          ,
          <article-title>Theoretical and Applied Principles of Information Technology for Supporting Medical Decision-Making Taking into Account the Legal Basis</article-title>
          ,
          <source>CEUR-WS</source>
          <volume>3038</volume>
          (
          <year>2021</year>
          )
          <fpage>172</fpage>
          -
          <lpage>181</lpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <given-names>T.</given-names>
            <surname>Hovorushchenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Herts</surname>
          </string-name>
          ,
          <string-name>
            <given-names>Y.</given-names>
            <surname>Hnatchuk</surname>
          </string-name>
          ,
          <string-name>
            <given-names>O.</given-names>
            <surname>Sachenko</surname>
          </string-name>
          ,
          <article-title>Supporting the Decision-Making About the Possibility of Donation and Transplantation Based on Civil Law Grounds</article-title>
          ,
          <source>in: Advances in Intelligent Systems and Computing</source>
          , Springer International Publishing, Cham,
          <year>2020</year>
          , с.
          <fpage>357</fpage>
          -
          <lpage>376</lpage>
          . doi:
          <volume>10</volume>
          .1007/978-3-
          <fpage>030</fpage>
          -54215-3_
          <fpage>23</fpage>
          .
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <given-names>Y.</given-names>
            <surname>Hnatchuk</surname>
          </string-name>
          ,
          <string-name>
            <given-names>T.</given-names>
            <surname>Hovorushchenko</surname>
          </string-name>
          ,
          <string-name>
            <given-names>O.</given-names>
            <surname>Pavlova</surname>
          </string-name>
          ,
          <article-title>Methodology for the development and application of clinical decisions support information technologies with consideration of civillegal grounds</article-title>
          ,
          <source>Radioelectron. Comput. Syst. № 1</source>
          (
          <year>2023</year>
          )
          <fpage>33</fpage>
          -
          <lpage>44</lpage>
          . doi:
          <volume>10</volume>
          .32620/reks.
          <year>2023</year>
          .
          <volume>1</volume>
          .03.
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <given-names>M.</given-names>
            <surname>Szostak</surname>
          </string-name>
          ,
          <article-title>Impact of gender differences in perception of creative identities of artist, creator, manager, entrepreneur and leader on sustainability</article-title>
          ,
          <source>Entrep. Sustain. Issues 9</source>
          .2 (
          <year>2021</year>
          )
          <fpage>10</fpage>
          -
          <lpage>36</lpage>
          . doi:
          <volume>10</volume>
          .9770/jesi.
          <year>2021</year>
          .
          <volume>9</volume>
          .
          <issue>2</issue>
          (
          <issue>1</issue>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref11">
        <mixed-citation>
          [11]
          <string-name>
            <given-names>K.</given-names>
            <surname>Ginis</surname>
          </string-name>
          ,
          <string-name>
            <given-names>S. E.</given-names>
            <surname>Stewart</surname>
          </string-name>
          ,
          <string-name>
            <given-names>L.</given-names>
            <surname>Kronborg</surname>
          </string-name>
          ,
          <article-title>Gender and Artistic Creativity: The Perspectives and Experiences of Eminent Female Visual Artists</article-title>
          ,
          <string-name>
            <surname>J. Creative Behav.</surname>
          </string-name>
          (
          <year>2023</year>
          ). doi:
          <volume>10</volume>
          .1002/jocb.605.
        </mixed-citation>
      </ref>
      <ref id="ref12">
        <mixed-citation>
          [12]
          <string-name>
            <given-names>M.</given-names>
            <surname>Szostak</surname>
          </string-name>
          ,
          <article-title>Gender differences regarding participation form in the arts receiving process. Consequences for aesthetic situation management</article-title>
          ,
          <source>Int. J. Contemp. Manag</source>
          . (
          <year>2022</year>
          ). doi:
          <volume>10</volume>
          .2478/ijcm-2022-0010.
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>