Search

Research in social media marketing studies ways to increase customers’ engagement with brand-generated social media posts. This can either be done through experiments, or corpus studies of existing social media posts. Experiments have the advantage that they are controlled, but they often lack ecological validity, while for corpus studies the reverse is often true. As a case study, we construct a corpus of 1761 brand-generated Instagram posts, looking at the effect of social presence (the perception of human contact) on different engagement metrics (likes and comments), taking the effect of possible confounds (theme of slogans, funniness, time) into account. We show how social media posts can be analyzed at different levels of granularity, to establish the strength of the effect of social presence. We hope that our work will help others to isolate the impact of different variables on post engagement on social media.

We reflect on user privacy concerns, transparency and informed consent for long-term interactions with personalized conversational agents. We argue that the common practice of asking users to sign an informed consent form is insufficient to accommodate the privacy concerns of the user. We propose that long-term engaging personalized conversational agents must include an explicit mechanism in their conversations to allow users to have better control over their stored personal information and to have transparency about who is allowed to view the stored personal information.

In this paper we discuss our road map to design a Dutch spoken conversational agent that helps older adults with self-management of their well-being. We conducted three usability studies with our conversational agent, each time after discussing it with different stakeholders. We discuss the challenges incorporating each stakeholder's wishes and needs whilst iteratively designing our agent.

Talking about odors and flavors is difficult for most people, yet experts appear to be able to convey critical information about wines in their reviews. This seems to be a contradiction, and wine expert descriptions are frequently received with criticism. Here, we propose a method for probing the language of wine reviews, and thus offer a means to enhance current vocabularies, and as a by-product question the general assumption that wine reviews are gibberish. By means of two different quantitative analyses—support vector machines for classification and Termhood analysis—on a corpus of online wine reviews, we tested whether wine reviews are written in a consistent manner, and thus may be considered informative; and whether reviews feature domain-specific language. First, a classification paradigm was trained on wine reviews from one set of authors for which the color, grape variety, and origin of a wine were known, and subsequently tested on data from a new author. This analysis revealed that, regardless of individual differences in vocabulary preferences, color and grape variety were predicted with high accuracy. Second, using Termhood as a measure of how words are used in wine reviews in a domain-specific manner compared to other genres in English, a list of 146 wine-specific terms was uncovered. These words were compared to existing lists of wine vocabulary that are currently used to train experts. Some overlap was observed, but there were also gaps revealed in the extant lists, suggesting these lists could be improved by our automatic analysis.

An important objective in health-technology is the ability to gather information about people's well-being. Structured interviews can be used to obtain this information, but these are time-consuming and not scalable. Questionnaires provide an alternative way to extract such information, yet they typically lack depth. In this paper, we present our first prototype of the Behaviour-based Language-Interactive Speaking Systems (BLISS), an artificial intelligent agent which intends to automatically discover what makes people happy and healthy. The goal of BLISS is to understand the motivations behind people's happiness by conducting a personalized spoken dialogue based on a happiness model. We built our first prototype of the model to collect 55 spoken dialogues, in which the BLISS agent asked questions to users about their happiness and well-being. Apart from a description of the BLISS architecture, we also provide details about our dataset, which contains mentions of over 120 activities and 100 motivations and is made available for usage.

It is widely held that smells and flavors are impossible to put into words. In this paper we test this claim by seeking predictive patterns in wine reviews, which ostensibly aim to provide guides to perceptual content. Wine reviews have previously been critiqued as random and meaningless. We collected an English corpus of wine reviews with their structured metadata, and applied machine learning techniques to automatically predict the wine's color, grape variety, and country of origin. To train the three supervised classifiers, three different information sources were incorporated: lexical bag-of-words features, domain-specific terminology features, and semantic word embedding features. In addition, using regression analysis we investigated basic review properties, i.e., review length, average word length, and their relationship to the scalar values of price and review score. Our results show that wine experts do share a common vocabulary to describe wines and they use this in a consistent way, which makes it possible to automatically predict wine characteristics based on the review text alone. This means that odors and flavors may be more expressible in language than typically acknowledged.

It is widely held that smells and flavors are impossible to put into words. In this paper we test this claim by seeking predictive patterns in wine reviews, which ostensibly aim to provide guides to perceptual content. Wine reviews have previously been critiqued as random and meaningless. We collected an English corpus of wine reviews with their structured metadata, and applied machine learning techniques to automatically predict the wine's color, grape variety, and country of origin. To train the three supervised classifiers, three different information sources were incorporated: lexical bag-of-words features, domain-specific terminology features, and semantic word embedding features. In addition, using regression analysis we investigated basic review properties, i.e., review length, average word length, and their relationship to the scalar values of price and review score. Our results show that wine experts do share a common vocabulary to describe wines and they use this in a consistent way, which makes it possible to automatically predict wine characteristics based on the review text alone. This means that odors and flavors may be more expressible in language than typically acknowledged.

This work presents a comparative study between two different approaches to build an automatic classification system for Modalityvalues in the Portuguese language. One approach uses a single multi-class classifier with the full dataset that includes eleven modal verbs; the other builds different classifiers, one for each verb. The performance is measured using precision, recall and F1. Due to the unbalanced nature of the dataset a weighted average approach was calculated for each metric. We use support vector machines as ourclassifier and experimented with various SVM kernels to find the optimal classifier for the task at hand. We experimented with several different types of feature attributes representing parse tree information and compare these complex feature representation against a simple bag-of-words feature representation as baseline. The best obtained F1values are above 0.60 and from the results it is possible to conclude that there is no significant difference between both approaches.