Cultural Heritage Topics in Online Queries: A Comparison between English- and Polish-Speaking Internet Users

Abstract

New communication channels and methods for retrieving information can provide increasingly precise data describing how cultural heritage is perceived, protected, promoted, and shared. Many internet users search for cultural-heritage-related topics using online search engines and databases. The purpose of this study was to identify the context and frequency of cultural heritage topics in Google search engine queries. The primary hypothesis was that keywords used in Google searches related to cultural heritage would be much more diversified in English than in Polish, perhaps because Polish has many fewer internet users than English. The keywords were selected because of their frequent use in the research literature, popular science publications, and central and local government strategy documents. The research employed selected online applications. A total of 3690 keywords were collected, with 1634 (44.3%) in Polish and 2056 (55.7%) in English. The numbers of retrieved keywords were similar for all research tools, but an in-depth analysis identified certain differences in the functions of the tools. The “cultural heritage” keyword in Polish (“dziedzictwo kulturowe”) was found mostly in national, regional, and local contexts. English queries included topics related to cultural heritage protection, conservation, restoration, and management and the financial value of cultural heritage. Some queries focused on world cultural heritage. The research shows that Polish-speaking users and English-speaking ones were interested in similar phrases. Therefore, the thought processes of internet users may be independent of their language.

1. Introduction

Cultural heritage components may originate from the natural environment or be manmade. They are most commonly divided into tangible, intangible, and natural categories [1,2]. However, the notion of “heritage” began to evolve in the final decades of the twentieth century; people began to apply the concept in place of such terms as “monument” or “cultural asset”. The international view of cultural heritage offers a global perspective which surpasses national, regional, or local ones. Cultural heritage assets harbour unique features that encourage communities to preserve and protect them. They can also evoke feelings and emotions [3]. Consequently, cultural heritage assets are no longer defined solely by their material attributes. This has given rise to an intangible cultural heritage, which has been long ignored in terms of protection [4]. Moreover, a new heritage category has recently come to light: digital cultural heritage [5].

Most people have positive associations with the concept of heritage and attribute positive values to it. Components of tangible culture, such as works of art, everyday objects, architecture, and landscapes are generally considered to be for the common good and beneficial to all, like intangible heritage components such as dances, music, theatre performances, rites, language, and memory [6]. This has led to the popular belief that investment in cultural heritage (and other forms of culture) is beneficial to the local economy not only in terms of cultural consumption, but also through increased employment and income [7,8].

Studies show that socioeconomic development should take into account cultural heritage and be based on the achievements of past generations [9,10]. Cultural heritage is a source of traditions, practices, and customs. Handicrafts, cuisines, and cultural events can be restored thanks to historical records and oral traditions. These can all can provide a framework for socioeconomic and cultural development. Nevertheless, cultural heritage components are fleeting and can be forgotten if no comprehensive effort is made to preserve them. Moreover, education and transgenerational transfer of traditional skills are necessary to achieve economic value through the use of heritage [11].

Living in today’s knowledge society implies a recognition of the importance of the past and consideration of cultural heritage as a fundamental background of our identities [12]. Information and communication technology offers easier access to and a more comprehensive view of cultural heritage artefacts. It can also further enrich and improve heritage education through innovative learning and teaching methods [13]. New communication channels and methods for retrieving information can provide increasingly precise data on how cultural heritage is perceived, protected, promoted, and shared, and also on the expectations of the audience [14]. Generally available search tools and social media can provide information on changing trends, moods, and opinions regarding cultural heritage from both the local and global perspectives [15,16].

An analysis of the contexts and frequencies of keywords can provide strategic marketing insights. An in-depth investigation of keywords used online can help to identify a market niche, find trends, and indicate future research directions [17]. Although many long-tail keywords can have complex links to a topic, their analysis and the use of copywriting consistent with analytical conclusions can improve target conversion [18]. It is not only the keywords used by search engine users (macro-level) that are analysed, but also the keywords used by consumers and then stored in the search logs of individual websites [19].

Many internet users search for cultural-heritage-related topics using online search engines and databases [20,21]. The purpose of this study was to identify the context and frequency of cultural heritage topics in Google search engine queries. The Google search engine was chosen due to its popularity [22]. The following research questions were posed:

Q1: In what context and how often did Google search engine users search for cultural heritage?

Q2: How diversified were cultural heritage queries in Google?

The primary hypothesis was that keywords used in Google searches related to cultural heritage would be much more diversified in English than in Polish, perhaps because Polish has many fewer internet users than English.

Internet content is most often collected for research purposes using advanced text data acquisition methods, such as text mining or web scraping [23,24]. Text exploration is an automated process whereby a natural language is processed to extract information from an unstructured text. It is employed in such scenarios as research efforts or searches for competitive advantage [25]. The tools described in the paper require no knowledge of programming techniques. This makes them useful to librarians, archival clerks, humanities researchers, and all those who handle content editing and metainformation [26]. The research followed the open-access concept by exploring the surface web, which consists of those resources available in the search index [27]. The tools we used can help to enhance descriptions of cultural heritage components. In this way, their availability (visibility) in search results can be improved for both humans (user experience; UX) and search algorithms (machine experience; MX). This is a complementary approach in line with the search engine content optimisation framework [28]. This paper shows how people search for cultural heritage components, which often shape the tourism potential of regions or countries. We further demonstrate how one can obtain information about user behaviour in the digital ecosystem and how this insight can be used in the process of shaping media messages, including those regarding the improved availability and diversity of cultural heritage information.

Although new topic detection is a well-researched field [29], our approach was unique. The research technique applied here involved the use of keyword research tools to find out the context and frequency of cultural heritage topics in online queries. We employed an uncommon approach within a rather atypical field related to cultural heritage, which is usually associated with tangible objects [30]. With the present research, we strive to debunk the perception of cultural heritage as consisting only of architectural sites, wayside shrines, religious structures, or environmental objects, and consider it in terms of search engine users and in the context of digital cultural heritage [31].

The remainder of this paper is structured as follows. Section 2 presents Polish research on verbal communication on the internet, research on English as a dominant language, a review of cultural heritage on the internet, and a definition of types of keywords and their impacts on search results, focusing on long-tail keywords. Section 3 presents the methodology: the research procedure and tools. Section 4 contains quantitative results, followed by semantic analysis results in Section 5. The paper ends with a summary.

2. Background

2.1. Language as Cultural Heritage

Every human behaviour intended to produce a response in others involves social communication, and language is a primary communication tool [32]. Communication is not only about facts but also involves interpretations, attitudes, values, and understandings and perceptions of reality. Therefore, language is not only an instrument of communication but an entire system that determines the process [33]. Therefore, language is the primary carrier of culture and intercultural communication. An individual using a certain language is, to some extent, a participant in the culture established by the group that speaks this language.

Every national language is a tool, material, and component of culture. It extends over the everyday lives of its users, their spirituality, and the entirety of their intellectual and artistic activity. Being the raw material of national literature, it preserves and reflects the complexity of historical experience. It constitutes the linguistic picture of the world, which is a collection of principles included in category-driven grammatical relationships and semantic lexical structures. It exhibits language-specific perceptions of individual components of the world and a general understanding of how it is organised, including its hierarchies and values. Therefore, every national language is rooted in the separate culture of the user community, which it affects [34].

Language is, in a way, a treasury and custodian of generations’ worth of knowledge about the world. This knowledge is in and of itself an important component of the cultural heritage of the community that speaks the language. The cognitive and cultural functions of a language are usually considered based on national (ethnic) languages. The linguistic differentiation of a population reflects cultural (and civilisational) differentiation, while cultural pluralism determines the abundance of cultural achievements of humanity. On the other hand, there is a strong need for the circulation of ideas and transfer of knowledge beyond the boundaries of national languages [35].

2.2. Introduction to Polish Research on Verbal Communication on the Internet

Polish is a large Central European language [36]. Polish language studies, Polish applied linguistics, and language pedagogy have identified “variations”, “dialects”, or “registers” of the language, with such qualifiers as “professional”, “expert”, “special”, “specialised”, or “specialist”. The use of specific phrases is most often associated with a specific circle, class, or social group that stands out from the population in terms of age or generation (language of the youth, students, or a generation), or profession (IT or legal jargon).

According to sociolinguists, a special language is usually limited to communities delimited by a kind of social bond, such as a class, stratum, circle, or professional group [37]. Moreover, Polish linguistics divides specialist Polish into a professional language and professiolect, which is mainly spoken, and academic language, which is primarily written. The same applies to various groups of enthusiasts, including those in grassroots retrogaming, retrocomputing, and general (digital) cultural heritage societies [38].

Polish research focusing on linguistic aspects of verbal communication on the internet began in the early days of the internet in Poland, in the late twentieth century. This research most often concerns models of communication on the internet, typology of internet communication, the ontological status of the language on the internet, analysis of the notions of text and hypertext, a genealogical map of the internet language, characterisation of individual linguistic subsystems in internet communication, etiquette in technology, use of various codes to construct a text to be published online, identification and analysis of plays on words and their potential impact on the Polish language, and implications of internet-based language and communication practices in the virtual space for everyday language and communication [39]. Such analyses may be relevant to the moulding of the media discourse on cultural heritage. They may be particularly useful for the promotion of local heritage, which is hard to promote to a large online audience.

2.3. English as a Dominant Language

Cultural diversity is inherent to linguistic diversity. In the time of globalisation and the global village, it is the most valuable resource of humanity, a treasury of traditions, and a reservoir of growth opportunities [40]. This assumption emphasizes the peaceful partnership collaboration of nations in mutual respect for their differences and individualities. It also attributes particular humanistic values to multilingualism and polyculturalism [41]. On the other hand, the use of multiple languages may be advised against in some domains. The roles of dominant languages in science and technology, economy, tourism, transport and logistics, politics, and particularly mass media are growing, as they facilitate international communication and make information available to a broader audience.

In general, global linguistic communication has always been multifaceted and diversified. However, some areas of social life exhibit a strong trend towards universalisation and unification, mostly in favour of English today. Its dominance is clear in science, where linguistic diversity is perceived as a hindrance to the transfer and perception of information [42]. English is the primary tool of international scientific communication today in terms of original works, citations, and researcher collaboration [43]. Nevertheless, expert knowledge and familiarity with the terminology used in a specific field are just as important [44].

The universal use of English as the language of international academic discourse breaches the barrier of multilingualism in science to facilitate the transfer of knowledge. Paradoxically, the ever-growing dominance of this language is sometimes considered a barrier in and of itself; it may significantly hinder the participation of researchers from outside English-speaking countries in the global scientific effort, limiting the impact of contributions in languages other than English on global scientific development [41]. The same applies to the flow of information more generally, particularly information of local relevance that one might wish to promote internationally, which could be the case for tourism or the promotion of local cultural heritage.

2.4. Cultural Heritage on the Internet

Online cultural content is available in many forms, such as texts, images, soundtracks, videos, and NFTs [45]. These items concern various topics, including art, handicraft, etc., and are written in various languages. Such content comes from diverse independent preservation organisations, such as museums, archives, libraries, or individuals, and is targeted both at amateurs and experts. The problem of finding and connecting information in such an environment of heterogeneous content delivery and data format can be both a barrier for end users trying to access cultural content and a challenge for content producers [46]. Recent decades have seen new bold campaigns by public institutions and private organisations to digitise cultural artefacts, yielding huge digital collections. These campaigns offer public access to millions of digital objects from many collections of cultural heritage through multilingual graphic user interfaces. However, large datasets such as digital libraries suffer from low availability to the general public and a difficult search process [47]. Smart conversational agents facilitate access to information in semantic networks through natural-language interaction and structured answers to user queries. Nevertheless, these tools are not as commonplace as search engines. In traditional portals, search is usually based on free text search (e.g., Google), database queries, and/or a stable classification hierarchy. Semantic content makes it possible to provide the end user with more “intelligent” facilities based on ontological concepts and structures [46].

2.5. Keyword Analyses

Research paper keyword analyses usually focus on one or multiple journals over a specific period. The frequency of keywords and their links to individual articles facilitate the visualisation of the results [48]. A keyword analysis can demonstrate whether a specific field is gaining popularity in the literature by arousing interest among researchers (a popular topic) or is relatively recent and paves the way for future efforts (an emerging topic) [49,50]. Such insight can hardly be gained through a traditional literature review [51].

Research paper keywords can be categorised according to two main attributes: dominance and perseverance. The former reflects the frequency of the keyword in a set of papers. It is the number of papers in which specific keywords occur. The latter attribute is related to the temporal continuity of the topic. Combined, the two attributes create a matrix of four areas. Each defines a homogeneous group of keywords [52]. Analysis of research paper keywords can identify descriptors of primary research topics and gene words that indicate knowledge domains formed through cross-referencing and hybridisation of core keywords [53].

If the number of papers or other sources is small, it is possible to analyse them manually using test and computing tools. For larger numbers of publications or online sources, more advanced methods for obtaining textual data are used, such as text mining or web scraping.

Web scraping is a technique for extracting information from online sources. It replaces the manual, mechanical, and repetitive feeding of a spreadsheet or other software. It is broadly used to analyse large datasets via various packages and scripts, most often in R and Python. It has been applied to analyse manager job postings [54], to analyse emotions during the COVID-19 pandemic [55], and in numerous other focus papers [56].

Packages in R used to harvest and explore large textual datasets from the internet include rvest [57] and downloader [58]. Beautiful Soup and Reguests-HTML [59,60] are popular Python packages. Social media data can be scraped using facebook-scraper and Selenium. The application programming interface (API), which facilitates interactions among applications, is an important factor in sharing online resources. Today, its most common context is Web API and REST API [61]. These tools can be used according to service terms and conditions. Slightly less commonly used tools for such analyses are keyword planners, which can identify and discover new keywords for a topic [22].

2.6. The Information Potential of Keywords

A keyword is a phrase of one or several words that search engine users use. Keywords identify specific content. They can be included in the metadata and help to better adjust search results to user queries. There are three primary types of keyword: branded keywords (such as names of brands), generic keywords (most often consisting of one or two words that do not designate the searched product or piece of information precisely, for example, “cultural heritage”), and long-tail keywords, which are detailed queries typed into a search engine (such as “rural cultural heritage”). Geolocated or geotargeted keywords with place or street names, for example, are becoming increasingly popular [62].

The foundation of the long-tail strategy is that a broad and diversified set of keyphrases that are niche, less competitive, and each able to individually generate limited traffic results in significant aggregated organic traffic [63]. Nonstandard, unpopular keywords and detailed, precise, phrases of multiple words increase a website’s reach if used in large numbers.

The idea of long-tail keywords was born from user behaviour. People increasingly search for specific information using complex queries of several words in various orders and diverse grammatical forms. According to the long-tail concept, users who type in uncommon, nonstandard, or long queries are searching for specific results and are more determined, making target conversion more probable [64].

Knowledge of keywords used online can help adjust content to users’ expectations. When the context and frequency of keywords are known, the content can be made more appealing to users. Consequently, they can reach specific websites more often. These activities are part of content marketing and search engine marketing (SEM), and are the basis of search engine optimisation (SEO) [65].

The context of specific keywords in users’ queries can be identified using online applications. Tools that propose keywords (keyword planners or generators) are based on keywords entered by users who search for information using search engines [66]. For example, when one enters a query into the Google search engine, it suggests other similar keywords. Keyword-suggesting tools fetch the suggestions from search engines and display them in their interface. However, the information is merely illustrative and the keyword frequency is not given in absolute values.

3. Materials and Methods

3.1. The Research Approach

Europe has no language that can stand for its identity as a whole and provide the emotional canvas for all its citizens [67]. According to the official position of the European Union, its linguistic symbol is multilingualism. The use of English in EU institutions for internal, member state, and external communication is growing for pragmatic reasons, which means that English is being promoted as the de facto EU communication language [68]. The spread of English is a sign of globalisation. It connects Europe and the rest of the world, while integrating EU states [69]. The link between language and identity in Europe consists of the fact that the use of English and other communication languages, and the interpreting and translating mechanisms in the EU, foster the circulation of information and ideas to strengthen the European identity [67]. This fact is of importance for the identification, collection, and promotion of cultural heritage, which is often local and can reach a broader audience only through descriptions and presentations in such global languages as English [70,71,72].

The research reference framework was linguistic globalisation. Globalisation within a language (linguistic globalisation) refers to all trends, modifications, and changes within a national modern language that result from homogenisation, simplification, and standardisation in a modern culture that is significantly affected by the media [73,74]. Such processes lead to a high level of predictability of linguistic communication that is apparent particularly in the acts of speech, speech genres, and discourses people employ. This approach presents two main aspects of linguistic globalisation: (1) as the use of a language in international communication to build understanding and (2) as the diffusion of components of one language into other languages [73,75]. This study also followed the framework of media linguistics. Media linguistics refers to the investigation of media-driven linguistic phenomena. These are linguistic events which are defined in shape and nature by individual media, taking into account the technological, social, economic, and cultural context [76]. Online linguistic research often faces questions of what and how to research. “New media” communication poses new theoretical and practical problems for researchers. Theoretical difficulties (terminology, missing analytical categories, typology issues) and practical obstacles (lack of existing models for describing and explaining phenomena) may prove to be a barrier [77].

The popularisation of the internet, and especially of user-generated content, affects contemporary linguistic behaviour [78]. Communication on the internet reinforces the hegemony of English and pushes Anglicisms into national languages, including Polish. On the other hand, the internet and social media may be spaces where regional or endangered languages can potentially be protected and restored [79]. Another theoretical pillar was media studies. This approach proposes the inclusion of linguistics methods in media studies because of the growing breadth and depth of media-mediated communication [74]. Modern linguistics can identify the language used in media at three levels: (1) the discourse: discourse analysis explores the roles of sender and receiver in discourse and the conditions set by technology and the medium; (2) the genre of media messaging; and (3) the text, which seems to be the favourite subject of linguists researching media. It is here that the most basic fact of linguistic globalisation takes place: the internationalisation of the dictionary [74].

3.2. Conclusion Mechanisms and Data Analysis

The present study employed both quantitative and qualitative methods. The data triangulation used here allows a phenomenon to be described using data from various sources to establish relationships between the analysed facts [80,81]. The combination of this triangulation and qualitative and quantitative approaches yielded exhaustive knowledge of the subject despite any differences and contradictions between the methods.

Quantitative research tackles the magnitudes of phenomena, respondent opinions, or the popularity of an opinion. These methods are based on quantifiable attributes (measurable, quantifiable, tangible features), usually expressed as numbers, letters, or graphics (charts and indicators). Quantitative research is often used to standardise data collection and allow generalisation of results. Quantitative projects aim to identify quantitative relationships between series of observations, focusing on the precision of the variable determination and measurement logic [82]. Moreover, quantitative research is often deductive. First, a phenomenon is assumed a priori, and is later confirmed through a search of data and information. A quantitative researcher asks “how many/much?”, because numbers are inherent to this approach [83].

Quantity is understood mostly in terms of the number of objects, while quality relates to either their positive evaluation or their evaluation in general (positive or negative). In this approach, quantitative research involves a quantitative (numerical) representation of the intensity of a phenomenon, while the qualitative view involves an in-depth analysis of any numerical values, particularly from case studies, and the presentation of potential interpretations [83].

One of the assumptions is that the researcher participates in the research. Personal experience affects the research process, and the interpretation of results is subjective. In this way, the researcher moves within a specific society’s or community’s culture and imparts a personalised sense and meaning to observed phenomena.

The conclusions are based on causality and descriptive analyses of the collected data. The employed approach involved the acquisition and aggregation of surface web data [84], followed by data visualisation and interpretation. Such research can be reproduced in other cultural contexts and with various respondent populations.

The literature offers many approaches for identifying the most representative keywords related to a specific phenomenon. Still, most previous works have focused on the use of statistics, syntax, grammar, or network-based characteristics to select representative keywords to analyse the target domain [85].

3.3. Research Tools

The interest in cultural heritage can be measured in the digital ecosystem (by counting specific queries, for example). It is also possible to identify the context in which cultural heritage elements can be found online. It can be approached from the point of view of content creators (what information is published and in what form) and from that of people interested in cultural heritage (what information is searched for, how, using what sources/channels, etc.). Keyword planners are tools that can reveal the context of cultural heritage topics in search engine queries to a certain extent.

The present study involved identifying the context and frequency of use of selected keywords in Google search engine user queries. The research employed three keywords in Polish and their counterparts in English: (1) PL: “dziedzictwo kulturowe”; EN: “cultural heritage”, (2) PL: “cyfrowe dziedzictwo kulturowe”; EN: “digital cultural heritage”; and (3) PL: “cyfrowe artefakty”; EN: “digital artefacts” (

Table 1. Definitions of the keywords used in the study.

Item	Keyword in PL/ENG	Definition	Reference
1	Dziedzictwo kulturowe/cultural heritage *	Heritage is the cultural legacy which we receive from the past, which we live in the present, and which we will pass on to future generations.	[89]
2	Cyfrowe dziedzictwo kulturowe/digital cultural heritage **	The digital heritage consists of unique resources of human knowledge and expression. It embraces cultural, educational, scientific, and administrative resources, as well as technical, legal, medical, and other kinds of information created digitally, or converted into digital form from existing analogue resources.	[90,91]
3	Cyfrowe artefakty/digital artefacts ***	Digital artefacts may be accessible from their beginnings in an electronic format (born digital) or may be a result of digitisation, i.e., a digital representation, replica, or digital substitute. Digital substitutes can take on the form of (faithful) virtual representations (3D), or may be digital phantoms or only an ersatz of the original (a so-called digital proxy), e.g., small illustrative graphics.	[10,92]

* Tangible culture, such as works of art and everyday objects, architecture, or landscapes, and intangible culture, such as dancing or theatre performances, traditional customs and rituals, language, and human memory are generally believed to be for the common good. Combined, they constitute the identity and integrity of local communities. ** According to UNESCO’s Charter for the Preservation of Digital Heritage, Digital Heritage comprises computer-based materials of enduring value, including texts, databases, still and moving images, audio, graphics, software, and web pages, in a wide and growing range of formats. Elements are frequently ephemeral and require purposeful production, maintenance, and management to be preserved. This heritage may exist in any language, in any part of the world, and any area of human knowledge or expression [93]. *** An artefact is an object that is a product of the human mind and human effort. A cultural artefact is a purposefully produced or transformed object to which a human has assigned a specific form and purpose. Artefacts comprise objects, both artistic and functional, existing in real and virtual space. An artefact can be either an analogue or a digital product [10].

). The results were juxtaposed. The analysis focused on keywords in Polish, as it is our national language, and English as a lingua franca employed regardless of the user’s nationality or background and one of the most commonly used languages in the world according to Ethnologue, a research centre for language intelligence [86]. These phrases were selected because they are common in research and popular science literature and central and local government strategic documents [87]. Furthermore, the role of cultural heritage in geopolitics, spatial, cultural, and socioeconomic development in the world is growing [9,88]. The present research focused on digital heritage, which is growing more popular [31].

We used selected online applications: (1) Keyword Tool (available at: https://keywordtool.io, accessed on 20 January 2023), (2) Kparser (available at: https://app.kparser.com, accessed on 20 January 2023), (3) SISTRIX Keyword Tool (available at: https://app.sistrix.com/en/keyword-tool/, accessed on 20 January 2023), and (4) Keyword Sheeter (available at: https://keywordsheeter.com/, accessed on 20 January 2023). These tools are employed in keyword research. They can be used to extract keyword lists from long-tail phrases [94,95]. We used the following search configuration: (1) Search term or keyword: (a) “dziedzictwo kulturowe”, (b) “cyfrowe dziedzictwo kulturowe”, (c) “cyfrowe artefakty”; (2) query language and country: (a) Polish (pl)/Poland (pl), (b) English (en)/United States (us).

Each tool yielded a database built for the keywords. Every database consisted of a set of keywords associated with the initial query. We then checked the seasonal variations of keyword frequency using Google Trends. Google Trends provides an overview of the popularity of a keyword in a specific period, which helps to identify any seasonal variations. Google Trends charts show keyword popularity on a scale from 0 to 100. The highest value, 100, indicates the greatest popularity in the period.

Keyword Tool is a free online keyword research instrument that uses Google Autocomplete to generate relevant long-tail keywords for any topic. The search terms suggested by Google Autocomplete are selected based on multiple factors. One of them is how often users searched for a particular search term in the past. Kparser is a professional keyword suggestion tool.

SISTRIX Keyword Tool helps to search for keywords by analysing queries users sent to search engines. In turn, Keyword Sheeter is a collection of tools for advanced keyword research. These tools can be used to answer the question of what keywords are used to search for specific content—in this case, related to the “cultural heritage” keyword. Such research tools can be used to identify new trends, popular phrases, tendencies, and new directions of change and development.

4. Results

Keyword Tool recorded 210 unique keywords, 29 question keywords, and 63 preposition queries for “cultural heritage” (in Polish). The keyword “digital cultural heritage” (in Polish) yielded 45 unique keyword suggestions, 25 questions, and 22 prepositions. The keyword “digital artefacts” (in Polish) yielded 56 unique keyword suggestions, 11 questions, and 43 prepositions.

Keyword Tool found slightly more queries in English, most of which (68%;

Table 2. The number of unique keyphrases according to Keyword Tool in the Polish and global internet.

Item	Keyword Tool 1	Keyphrases
		Cultural Heritage		Digital Cultural Heritage		Digital Artefacts
		PL	EN	PL	EN	PL	EN
1.	Keyword suggestions	210	309	45	54	56	82
2.	Questions	29	84	25	3	11	11
3.	Prepositions	63	121	22	15	43	24
Total		302	514	92	72	110	117

PL—queries in Polish; EN—queries in English. ¹ https://keywordtool.io (accessed on 17 February 2022).

) related to “digital heritage” (in English). Keyword Tool found 1207 unique keywords for the analysed keywords in total.

We identified a total of 1567 keywords related to the analysed queries using Kparser: “cultural heritage” (EN) yielded 962 keywords (61.4%); “digital cultural heritage” (EN) yielded 342 keywords (21.8%); and “digital artefacts” (EN) yielded 98 keywords (16.8%). SISTRIX Keyword Tool identified just as many keywords (

Table 3. The number of unique keyphrases according to Kparser and SISTRIX Keyword Tool.

Keyphrase	Online Application
	Kparser 1		SISTRIX Keyword Tool 2
	PL	EN	PL	EN
cultural heritage	165	409	200	188
digital cultural heritage	18	91	100	133
digital artefacts	14	84	56	109
Total	197	584	356	430

PL—queries in Polish; EN—queries in English, ¹ Kparser: https://app.kparser.com (accessed on 18 February 2022), ² SISTRIX Keyword Tool: https://app.sistrix.com/en/keyword-tool (accessed on 18 February 2022).

).

We recorded a total of 916 keywords linked to the topic of cultural heritage using Keyword Sheeter. Google users most often searched for “cultural heritage”. The phrase “digital artefacts” was used less often and in more diversified contexts (

Table 4. The number of unique keyphrases according to Keyword Sheeter.

Keyphrases 1	PL		EN
	Number of Keyphrases	Percentage (%)	Number of Keyphrases	Percentage (%)
cultural heritage	192	33.3	204	60.2
digital cultural heritage	196	34.0	49	14.5
digital artefacts	189	32.8	86	25.4
Total	577	100	339	100

PL—queries in Polish; EN—queries in English, ¹ https://keywordsheeter.com/ (search on 17 February 2022).

).

Keyword Tool offered the most keywords, with 32.7% of the set. However, aggregate counts were not overly varied (

Table 5. The total number of queries by application and language.

Keyphrases		Cultural Heritage		Digital Cultural Heritage		Digital Artefacts		Total	Percentage (%)
Item	Query Language	PL	EN	PL	EN	PL	EN
1.	Keyword Tool	302	514	92	72	110	117	1207	32.7
2.	Kparser	165	409	18	91	14	84	781	21.2
3.	SISTRIX Keyword Tool	200	188	100	133	56	109	786	21.3
4.	Keyword Sheeter	192	204	196	49	189	86	916	24.8
Total		859	1315	406	345	369	396	3690	100

). The tools combined collected a total of 3690 keywords, with 1634 (44.3%) in Polish and 2056 (55.7%) in English.

The numbers of retrieved keywords were similar for all the research tools, but an in-depth statistics inquiry identified certain differences in how each tool worked. Three of them provided more keywords for queries in English. Only one, Keyword Sheeter, offered more example queries in Polish (Figure 1).

According to Google Trends, the English keyword “cultural heritage” was much more common than its Polish counterpart, “dziedzictwo kulturowe”. Queries concerning cultural heritage in English were recorded in most countries in the world in 2021, but the keyword “dziedzictwo kulturowe”, written in Polish, was found only in queries from Poland. The other keywords exhibited similar trends, with “digital heritage” in English being much more prevalent than “cyfrowe dziedzictwo” in Polish.

5. Discussion

It is not uncommon for cultural heritage to determine the tourist potential of a region or country. Note that cultural heritage covers not only tangible and intangible components, but also natural (environmental) and digital assets. Some of these components are (digitally) represented in the digital ecosystem, which means they can be found online on websites or in internet databases. However, the accessibility of such content in search results is affected by multiple factors. These include proper textual descriptions with keywords used by people searching for information online. The identification of these keywords and their use in descriptions of cultural heritage assets may improve the assets’ ranks on search engine results pages, which would improve their chances of reaching a broader audience. Many researchers to date have focused on deep web analysis, machine learning, and exploration of large datasets using algorithms and programming techniques [96,97,98,99]. This study investigated the surface web and information available through general tools with a graphic user interface. This is a slightly different approach, which can be used to rapidly obtain small information samples useful at a particular moment.

5.1. The Many Sides of Keyword Analysis

Many researchers have assessed the scale or intensity of specific phenomena, taking into account such factors as geolocation, context, frequency, and communicated sentiments. Google Trends has already been used to analyse users’ interests across various fields. However, the purpose of big data utilisation is now shifting from monitoring to forecasting. Still, accurate forecasting will require additional analyses, such as of content and sentiment [100]. For example, it was demonstrated that Google Search volume has an additional predictive power regarding energy price volatility. The volume of Google searches for energy-related keywords is a significant predictor of energy commodity pricing volatility with incremental predictive power. Keyword planners have been successfully used in disease and disorder prevalence analyses. A Google Trends analysis unveiled an increase in searches for insomnia during the COVID-19 pandemic [101]. Moreover, data mining algorithms can be used to predict the trends of outbreaks. Predictions might support policymakers and healthcare managers in planning and allocating healthcare resources [102]. Pavelea and Nisioi [103] analysed Google search trends in four categories using Keyword Planner: travelling, cooking, fitness, and food delivery. They demonstrated that along with mental health, people’s everyday habits had been significantly impacted since the outbreak of the COVID-19 pandemic.

Keyword analysis can provide insight into domains that used to be investigated mainly using questionnaires, surveys, or interviews. According to Stephens-Davidowitz [104], estimates using Google search data are 1.5 to 3 times larger than survey-based estimates. Keyword analyses are used in economics studies as well. Huang and colleagues [105] evaluated the predictive capabilities of online search data. They proposed a hypothesis that although Google Trends is an important measuring tool for quantifying investor interest, any signals from changes in the search volumes depended on the sentiment regarding the searched phrases. Fan et al. [106] analysed the correlation between the search volume on Google Trends and the Taiwan weighted stock index. Their research revealed a correlation between the use of names of companies on the Taiwan 50 index as search keywords and the rise and fall of the TAIEX index. Huynh [107] analysed the impact of Google keywords on the number of new businesses (and amounts of capital registered) in Vietnam, a Southeast Asian country, after the Year of the Entrepreneur in 2016. Google Trends search query time series data have been used to improve forecast accuracy. Bangwayo-Skeete and Skeete [108] demonstrated that Google Trends insights offer significant advantages to forecasters, especially in tourism. Therefore, policymakers and business practitioners can use the forecasting capabilities of Google search data in planning. Böhme et al. [109] demonstrated how georeferenced online search data could be used to measure migration intentions in origin countries and to predict bilateral migration flows. Their study focused on the quality of search results and the potential to draw conclusions from data obtained from online search engines. The research was complemented by user behaviour analyses, which revealed that the impact of ICT on young people had been overestimated. Young people exhibit apparent ease of use and knowledge of digital devices. They rely on search engines to a large extent and browse rather than read content. They also lack the ability to assess online information critically and analytically [110]. Research has shown that finding information online is difficult and challenging because of the extremely large volume of data. A search engine can be used for this task, but it remains difficult to cover all the webpages on the internet [111].

The studies mentioned above offer an in-depth insight into (usually) long-term data from the deep web [112]. Our approach was different; it analysed the surface web “here and now”. Therefore, it can be referred to as a “snapshot of the existing state”. It does not claim to answer the questions of the reasons behind user behaviour or the foundations of search technology. We analysed the results of an ad hoc model. This approach offers a relatively quick insight into current trends and phenomena.

5.2. Search Query Data for Analysing Tourists’ Online Search Behaviour

Accurate forecasting of demand is vital for the success of tourism businesses [113]. With the growing importance of the internet for travel planning, understanding the online domain of tourism is increasingly important in the identification of potential solutions for the effective marketing of travel destinations [114]. Marketers need a deeper understanding of the behavioural aspect of search engine use [113,115]. Moreover, keyword analysis can be useful when forecasting tourism demand and the potential numbers of tourists in high and low seasons [116]. It can also be used to identify the most popular destinations.

Dergiades et al. [117] used an aggregate search engine volume index adapted to various languages and search engines to forecast the volume of foreign tourists. Xiang and Pan [115] discovered travel query patterns and commonalities and differences in travel queries concerning different cities by analysing transaction log files from several search engines. They demonstrated that the keywords in travellers’ queries reflected their knowledge about a city and its competitors. Siliverstovs and Wochner [118] established the context of tourism demand to be particularly useful. They found search-based tourism demand predictions to approximate reality rather accurately, on average. This suggests that search-based indicators may serve as valuable real-time complements to guide economic policy. Xiang et al. [114] analysed the visibility of the information about specific destinations, the visibility of various industry sectors within destinations, and the power structure of websites representing a specific destination. They revealed that although there was a large amount of information indexed, travellers could only access a minuscule part of the domain (surface web). Additionally, a relatively small number of websites dominated the search results.

The cited research demonstrates that the analysis of data from search engines can support tourism demand analysis in general, trend analysis, “snapshot” analysis, monitoring, predictions, and forecasting. The queries entered into search engines include millions of different searches by tourists. Not only do they reflect the trends of the searchers’ preferences for travel products, but they also offer predictions of their future travel behaviour [119]. The studies take into account that the queries are multilingual and search engines differ. Conclusions are sometimes supported by statistical analyses. This, however, can make any repetition of the analyses by marketers or interactive agencies and subsequent optimisation of a website (and/or content) time-consuming or even impossible for more complex models. Meanwhile, the methods proposed here, using universal and easily available automated tools, have a better chance of being used regularly to edit content to be published online.

5.3. Cultural Heritage in Online Queries

The present research provides insight into the context of cultural-heritage-related internet user queries (Q1: In what context and how often did Google search engine users search for cultural heritage?). The users searched for relationships between cultural heritage, tourism, and intellectual property and the links between “monument” and cultural heritage. They then looked for information about cultural heritage as a driver of local, regional, socioeconomic, and cultural development (Figure 2). This information can be used to prepare content for cultural-heritage-related websites. The data can also be used in research and popular science publications, tutorials, FAQs, graphics, infographics, or videos (thematic materials in general).

“Digital cultural heritage” was searched by internet users in the context of digital archives, library collections available in digital form or through digital means, digital museums, and digital works of art. Meanwhile, the keyword for “digital artefacts” in Polish, “cyfrowe artefakty”, was used in the general context of digitalisation rather than strictly concerning cultural heritage. Users searched for digital artefacts in phrases concerning digital images (graphics), availability of digital content, digital components of virtual worlds, and the quality of graphic files.

It is impossible to provide absolute numbers that would demonstrate how often cultural heritage topics occurred in queries posted in the Google search engine (under the employed research design). The tools employed in the research provide overviews and list the most common keywords from user queries. Although results from Google Trends can be a point of reference, they are rather synoptic as well. Moreover, various synthetic indices can reflect the popularity of a keyword, including Search Volume and Cost-Per-Click (CPC), as well as original measures (indices) based on the frequency range of each word from the phrase among the results.

The “cultural heritage” keyword in Polish (“dziedzictwo kulturowe”) was found mostly in national, regional, and local contexts. Internet users searched for a definition of cultural heritage and for examples of the cultural heritage of various ethnic groups, including Polish cultural heritage and that of Europe, Africa, South America, Belarus and Lithuania; such regions of Poland as Lesser Poland, Kashubia, Opole region, Podlachia, Podhale, Greater Poland, or Subcarpathia; China, Czechia, Greece, France, Greece and Rome, Upper Silesia, Cracow, Gdańsk; and even such municipalities as Żukowice or Kozienice. These queries concerned specific examples of the cultural heritage of localities, regions, or states.

The English keyword “cultural heritage” was found in a slightly broader range of contexts than the Polish keyword (Q2: How diversified were cultural heritage queries?). English queries included cultural heritage protection, conservation, restoration, and management and the financial value of cultural heritage. Some queries focused on world cultural heritage or the heritage of specific cities and regions, such as the Philippines, Nepal, India, Africa, Zimbabwe, Venezuela, Zambia, Japan, Tanzania, Kosovo, Bangladesh, and Zamboanga City, or of specific communities, such as the African Yoruba people. These queries could be somewhat exotic to a Polish internet user. Queries in both languages concerned cultural heritage in UNESCO documents and cultural heritage in legal and tourism contexts. Slightly less common queries asked about cultural heritage and food tourism, handicraft, or intangible heritage. However, it is worth noting that this context may be relevant to other keywords as well. Users also searched for publications on cultural heritage and related events, such as seminars and conferences.

6. Conclusions

The results were inconsistent with the hypothesis. We identified only slightly fewer keywords related to cultural heritage in Polish than in English. Note that the research tools did not provide information on the number of occurrences of individual keywords; they were recorded as single cases. This means that Polish-speaking users posted similar queries (keyphrases) to English-speaking ones. The train of thought of these internet users could be independent of their native language: cultural heritage information was searched using the same queries, possibly following the same trains of thought, but expressed in different languages.

The English-speaking audience is international and much larger than national user groups. Put simply, this means that the “Polish internet” is much smaller than the English-speaking internet community. Therefore, to win a larger audience for cultural heritage content, one should consider English and local language versions or just an English version. It may play a significant role in the education and promotion of cultural heritage.

Although all the tools retrieved similar volumes of keywords, an in-depth analysis demonstrated potentially significant differences in how each tool worked. This means that one should employ several alternative tools when searching for keywords used in search engines to build a more comprehensive and diversified database.

6.1. Theoretical Implications

According to the “long tail” concept, online commerce does not follow the principles of traditional trade. The sale of products that are less popular and yet available in a broad palette generates more revenue than the sale of the most popular products ordered in large volumes. This means that the income from niche products may exceed that from mass-produced commodities. In other words, the total value of niche markets is greater than that of the dominant mainstream market. Having a broad selection of products, the seller may generate a greater total revenue from individual, rarely searched items than the revenues for the most popular, large-scale products or services. The same concept applies to content marketing. Industry-specific text in which diverse phraseology is used can improve a website’s position in search results, in line with search engine content optimisation (SEO).

Knowledge of the profile of the target is just as important when promoting cultural heritage as it is in the online sale of products or services. Someone who shapes a media message might decide to use a persona, which is a profile of an imaginary person with a set of features typical of the content’s target. A persona can be useful when writing texts with targeted keywords and language adjusted to the perception of the intended audience. The keyword planners discussed in the paper can be used to search for diversified and uncommon combinations of keywords.

6.2. Practical Implications

The presented methods could be particularly useful in ad hoc search engine optimisation (SEO). SEO should not be based solely on the intuition of the marketer, copywriter, content auditor, or optimiser, or those charged with SEO. Their efforts should be founded on a synergistic combination of data analysis, expertise, observations, and intuition. Keyword analysis involves searching, selecting, and assessing words and their combinations, which requires knowledge of the frequency of their use, trends, variations, and benchmarking analyses. This means, in practice, that keyword analysis with an ad hoc model can be used to analyse trends, seasonal variations, and competitiveness of keywords to support a preplanned strategy. Keyword analyses are useful for creating guidelines for new content. This may be of particular importance for improving reach, which offers the potential to address a wider audience to better promote cultural heritage and drive tourism demand.

6.3. Limitations of the Research

The present work was quantitative research. Repetitions among the databases were not verified. It is possible that a given keyphrase occurred in all databases and was measured by every test tool. The Section 4 presents quantitative statistics, i.e., the number of all recorded keyphrases, not the number of unique keyphrases.

Suggestions generated by Google’s autosuggest function come from actual searches with the search engine and show trending keywords. They are also affected by the user’s location and previous searches. Moreover, the results are filtered by algorithms to remove suggestions in line with Google’s purposes. This fact does affect the results.

Keyword suggestion tools are most often used in content marketing. They allow the assembly of a set of keywords used by users, which may increase website traffic. These tools can also be used to unveil current trends by analysing search engine queries. However, this study was not focused on the marketing value and conversion potential of the keywords. Therefore, the set contained keywords that may be of limited use for cultural heritage website positioning.