Experiencing cultural heritage is a voyage of discovery and learning, where emerging insights and serendipity play a significant role. The experience also happens in a blended personal and social context. At the broader level, engagement is longitudinal - what we learn from modern cultural experiences (daily life in our surroundings) can provide clues to analogous interests in cultural materials, and vice versa. The richness of personal experience poses challenges and opportunities for capturing preferences in ways that support a user's experience with cultural heritage across institutions and over time, both in the digital realm and where digital interaction blends with a physical space.
It is important to recognize that the domain’s data itself is a moving target – available data (particularly linked open data) is emerging rapidly. And there are a growing number of initiatives for institutions to share and harmonize their data representations online, which means the potential to integrate information models and user profiles across cultural institutions will continue to evolve.
Copyright held by the author.
Our work in cultural heritage has been focused primarily on
the user experience and design of applications for museums
and archives, and helping institutions plan to incorporate
linked data to enhance people’s experience with their
cultural assets. We increasingly support institutions that
want to share and enrich data via federated approaches,
allowing information access to expand over time and across
institutional boundaries. Rich personalization is important.
Many tasks will need a greater level of support as the
volume of digital information – and associations via linked
data capabilities – grows in the coming years. As we design
interfaces, we explore the personal experiences people have
with cultural heritage information, whether online or in
person, and how their expectations and interactions with
cultural heritage evolve over time. The considerations that
arise from that work are the subject of this paper.
CONSIDERATIONS
Designing personalized interactions that are
companionable, flexible, and not awkward is vital. To
achieve that aim, personalization models need to provide a
great degree of user transparency [
Preferences Interact with User Knowledge At any one time, a person’s interaction with cultural heritage has a purposeful dimension – whether that is entertainment, knowledge-seeking for its own sake, or resource-seeking for some outside task.
To achieve a goal of supporting lifelong learning, it is important to focus on the general user experience with digital cultural heritage interactions. Yet there are specialized audiences who have their own needs for personalization. Scholars and practitioners in the field of cultural heritage have a driving motivation for discovering, interpreting, analyzing, synthesizing, publishing, and sharing. Their requirements are particularly rich, due to their specialized and detailed knowledge in particular subject areas. Educators have varying levels of experience and knowledge, and they act as surrogates for others who do not share their level of knowledge. An educator’s interaction with cultural heritage can be largely driven by a task that is focused outside themselves, as their role is primarily communication of cultural heritage information to others (although they also have their own personal interests to balance with their professional focus areas).
The richer a person’s experiences and knowledge, the richer
and more nuanced their personalized modeling may need to
be. And the broader the possible goals and tasks, the more
contextually aware applications need to become. At the
same time, single interactions between the individual and
an institution may be motivated by needs that are separate
from preferences and knowledge. For example, Falk [
Preferences have Scenarios and May be Transitory There are often situations where a person is engaging with cultural heritage as an aspect of a very specific, directed task. This could be writing a paper as a school student, or researching a book as a scholar, or preparing a treatment for a major artwork as a museum conservator. While a person will often be clear about their particular context when interacting with cultural heritage, they may not externalize that context to a supporting technology.
When that task is completed, whether in two weeks or two years, the intensity of focus decreases. In most cases there is still an interest in a particular subject or area of culture, but the goal that prompted a strong, focused interest may have waned. The corresponding strength of the preference may need to be tuned accordingly.
Preferences Emerge over Time When people encounter something for the first time, they may not immediately sense its significance. Tastes (and emotional connections to cultural objects or experiences) are emergent and are often recognized only on reflection, rather than “in the moment.” So something “stays with you” after the experience, or you recognize something as valuable/important only in the context of subsequent experiences with other things (they form a pattern that makes a whole, and a preference forms at that higher level). Strong interests from one interaction may over time have evolved or been eclipsed by more lasting connections with what were, at the time, weaker signals of engagement. In this way, personalization for cultural heritage may be more nuanced and longitudinal than preference and recommendation modeling required in other domains. When interacting with art and exhibitions, it is also helpful to recognize that the experience itself may be what is most important to the user, not necessarily the specific object of the experience. In a recent conversation, two museum-goers described in great detail an electronic exhibition piece where they interactively engaged with art. They described deep engagement with the experience but had trouble recalling the specific art that was the focus.
As systems collect data about interactions (whether clicks,
views, likes, saves, shares), it becomes important to discern
when and how to interpret interaction as interest, and assess
their actual longevity. It is useful to identify what elements
of personal interest/engagement are relevant to the activities
available at a particular cultural venue (whether this is a
formal cultural institution/site, an informal urban location,
or a purely digital online interaction), and identify ways to
interpret what someone “takes away” from the experience.
And algorithmic interpretation gets harder as the time gap
grows between the experience and the expression in an
electronic form. More distant value judgments (ratings,
suggestions from one user to another, etc.) can have lower
validity in recommendation algorithms [
Preferences Evolve Through Lifelong Experiences
Where do personal preference signals come from? Cultural
linked data used in education, media, tourism, gaming will
produce aspects of personal interest that can be reflected
back into cultural heritage preferences. This will require a
broader – and more nuanced – way of modeling
“preference”, and reflects the PATCH1 workshop goal of
exploring a longitudinal perspective that encompasses
lifelong learning. Various dimensions to support and
evaluate models have been proposed [
At the broader level, engagement crosses timespans – what
we learn from modern cultural experiences (daily life in our
surroundings) can perhaps translate to analogous interests
in cultural materials from other time periods [
Preferences are Balanced by Serendipity Personalization models need to guard against oversimplicity or rigidity, and in that way foster discovery and learning. At a basic human level, people seek to engage with culture and art in order to find delight in something new as well as to experience known and loved objects. Serendipity and discovery are a significant motivation for exploring cultural heritage. This is particularly true among scholars and curators, whose life work centers around interpretation and discovering new perspectives. It is also true for people who engage in conservation and built cultural environments (archaeologists, historians, preservationists, architects) who want to engage in the latest science and interpretation.
Some implementations of personalization can be restrictive,
even if the intention is to reduce informational “noise” and
increase relevance for a user. The “filter bubble” [
Preferences are Influenced by Social Interactions
When I go to a museum with other people, we engage with
things that Group/social interaction in physical space makes
it harder to know what is persistently preferential for the
individual rather than a reflection of an immediate social
group dynamic. What do I “keep for later” and transfer
between contexts, and what is purely “in the moment” for
my relationship with the people and place?
One research area to consider is exemplified by the
Epiphany Project [
In addition, the role of intra-group profiling of personalized
dimensions plays a role in weighting recommendations and
interactions in situations where the experiences are social.
Another aspect of social interaction with taste-making is
that a person’s knowledge of participants in a community
and the mutual alignment of interests can affect their
interpretation of how they judge recommendations [
Preferences are Uneven Across Descriptive Dimensions
The dimensions of description and interpretation of cultural
material and places are deeply multi-faceted. As we know
that not all dimensions have equal weight in a person’s
engagement with culture [
As we consider rich dimensions, it is important also to
consider the challenges that arise from such deepening data
pools. In the recent book Understanding Context [
Creating an emerging personal profile could involve
interacting with an agent that is focused on your
personalized experience; one that both guides and listens to
a person’s expressions of interest [
Focusing on the institution’s perspective, how does a
particular cultural environment access an individual’s
personalized model to align the person’s experience with
information or an activity? What permissions might be
required, and perhaps how would an automated agent be
empowered to provide that on a person’s behalf?
POSSIBLE ATTRIBUTES OF PERSONAL MODELS
Reflecting on the above considerations for personalization
leads to attributes that could be incorporated into
frameworks explored by PATCH participants, as well as
others in the cognitive computing and HCI communities.
• Longitudinal preference building: Both the elements of
preference (signals of interests expressed by an
individual) and their strengths may be accrued over time,
so the longevity of interest and the context in which it
arose is evaluated over time.
• Organic movement and decay: Individuals can remain
interested in things long after their focus or tastes have
changed. But models need to provide a method of
“decay” for interests that are not acted on, as preferences
change over time, and outdated preferences can be
perceived as noise by users.
• Recognize shared and social interactions: Models
could usefully identify social contexts that people are in
when preferences are engaged, and have ways of
realigning their weightings accordingly – or prompt the
user to take greater control of the experience.
• Allow dynamic weighting: Recognize context and user
expectations, and provide an appropriate level of control
for a person to express goals and needs. Then have those
expressions reflect back into the preference model.
• Provide simple frameworks for permission-giving:
Plan for an emerging ecosystem of information around
personal interests information and preferences. Identify
how to make the collection and use of information as
transparent as possible, to foster trust and communication
among the parties involved (whether humans, institutions,
or digital agents) [