Search results

The growth of social media and the sharing economy is generating abundant unstructured image and video data. Computer vision techniques can derive rich insights from unstructured data and can inform recommendations for increasing profits and consumer utility – if only the model outputs are interpretable enough to earn the trust of consumers and buy-in from companies. To build a foundation for understanding the importance of model interpretation in image analytics, the first section of this article reviews the existing work along three dimensions: the data type (image data vs. video data), model structure (feature-level vs. pixel-level), and primary application (to increase company profits vs. to maximize consumer utility). The second section discusses how the “black box” of pixel-level models leads to legal and ethical problems, but interpretability can be improved with eXplainable Artificial Intelligence (XAI) methods. We classify and review XAI methods based on transparency, the scope of interpretability (global vs. local), and model specificity (model-specific vs. model-agnostic); in marketing research, transparent, local, and model-agnostic methods are most common. The third section proposes three promising future research directions related to model interpretability: the economic value of augmented reality in 3D product tracking and visualization, field experiments to compare human judgments with the outputs of machine vision systems, and XAI methods to test strategies for mitigating algorithmic bias.

Divisia component data is used in the training of an Aggregate Feedforward Neural Network (AFFNN), a general-purpose connectionist system designed to assist with data mining activities. The neural network is able to learn the money-price relationship, defined as the relationships between the rate of growth of the money supply and inflation. Learned relationships are expressed in terms of an automatically generated series of human-readable and machine-executable rules, shown to meaningfully and accurately describe inflation in terms of the original values of the Divisia component dataset.

Cultural heritage archiving is experiencing an increase in digitalisations of artefacts in the last 15 years. The reason behind this trend is a demand for providing information about the artefact in a more accessible way to the audience, for example, through online delivery or virtual reality. Other reasons might be to simplify and automate the management of artefacts. Having a ‘digital copy’ of artefacts, allows one to search an archive and plan its storage and dissemination in a comprehensive manner. With the increased digitalisation comes an increased use of artificial intelligence [AI] applications. AI can be very beneficial in classifying artefacts automatically through machine learning [ML] and natural language processing [NLP]. For example, an algorithm can identify the source and age of artefacts based on an image and can do this much faster for a large collection of photos than a human. Although AI provides many benefits, it also presents challenges: Sophisticated AI techniques require certain insights on how they work, need specialists to customise a solution, and require an existing large dataset to train an algorithm. Another challenge is that typical AI techniques are regarded as black boxes, which means they decide, but it is not obvious why a decision has been made. This chapter describes a project in collaboration with the South Australian Museum [SAM] on the application of AI to extract material lists from a description of artefacts. A large dataset to train an algorithm did not exist, and hence, a customised approach was required. The outcome of the project was the application of NLP in combination with easy-to-customise rules that can be applied by non-IT specialists. The resulting prototype achieved the extraction of materials from a large list of artefacts within seconds and a flexible solution that can be applied on other collections in the future.

The World Bank report Changing Wealth of Nations 2018 is only the most recent reminder of how much poorer Africa is becoming, losing more than US$100 billion annually from minerals, oil, and gas extraction, according to (quite conservatively framed) environmentally sensitive adjustments of wealth. With popular opposition to socioeconomic, political, and ecological abuses rising rapidly in Africa, a robust debate may be useful: between those practicing anti-extractivist resistance, and those technocrats in states and international agencies who promote “ecological modernization” strategies. The latter typically aim to generate full-cost environmental accounting, and to do so they typically utilize market-related techniques to value, measure, and price nature. Between the grassroots and technocratic standpoints, a layer of Non-Governmental Organizations (NGOs) do not yet appear capable of grappling with anti-extractivist politics with either sufficient intellectual tools or political courage. They instead revert to easier terrains within ecological modernization: revenue transparency, project damage mitigation, Free Prior and Informed Consent (community consultation and permission), and other assimilationist reforms. More attention to political-economic and political-ecological trends – including the end of the commodity super-cycle, worsening climate change, financial turbulence and the potential end of a 40-year long globalization process – might assist anti-extractivist activists and NGO reformers alike. Both could then gravitate to broader, more effective ways of conceptualizing extraction and unequal ecological exchange, especially in Africa’s hardest hit and most extreme sites of devastation.

The increasing importance and proliferation of text data provide a unique opportunity and novel lens to study human communication across a myriad of business and marketing applications. For example, consumers compare and review products online, individuals interact with their voice assistants to search, shop, and express their needs, investors seek to extract signals from firms' press releases to improve their investment decisions, and firms analyze sales call transcripts to increase customer satisfaction and conversions. However, extracting meaningful information from unstructured text data is a nontrivial task. In this chapter, we review established natural language processing (NLP) methods for traditional tasks (e.g., LDA for topic modeling and lexicons for sentiment analysis and writing style extraction) and provide an outlook into the future of NLP in marketing, covering recent embedding-based approaches, pretrained language models, and transfer learning for novel tasks such as automated text generation and multi-modal representation learning. These emerging approaches allow the field to improve its ability to perform certain tasks that we have been using for more than a decade (e.g., text classification). But more importantly, they unlock entirely new types of tasks that bring about novel research opportunities (e.g., text summarization, and generative question answering). We conclude with a roadmap and research agenda for promising NLP applications in marketing and provide supplementary code examples to help interested scholars to explore opportunities related to NLP in marketing.

Companies in all industries are searching for new sources of competitive advantage since the competition in their marketplace is becoming increasingly intensive. The resource-based view of the firm explains the sources of sustainable competitive advantages. From a resource-based view perspective, relational based assets (i.e., the assets resulting from firm contacts in the marketplace) enable competitive advantage. The relational based assets examined in this work are brand image and corporate reputation, as components of brand equity, and customer value. This paper explores how they create value. Despite the relatively large amount of literature describing the benefits of firms in having strong brand equity and delivering customer value, no research validated the linkage of brand equity components, brand image, and corporate reputation, simultaneously in the customer value–customer loyalty chain. This work presents a model of testing these relationships in consumer goods, in a business-to-business context. The results demonstrate the differential roles of brand image and corporate reputation on perceived quality, customer value, and customer loyalty. Brand image influences the perception of quality of the products and the additional services, whereas corporate reputation actions beyond brand image, estimating the customer value and customer loyalty. The effects of corporate reputation are also validated on different samples. The results demonstrate the importance of managing brand equity facets, brand image, and corporate reputation since their differential impacts on perceived quality, customer value, and customer loyalty. The results also demonstrate that companies should not limit to invest only in brand image. Maintaining and enhancing corporate reputation can have a stronger impact on customer value and customer loyalty, and can create differential competitive advantage.

The expansion of marketing data is encouraging the growing use of deep learning (DL) in marketing. I summarize the intuition behind deep learning and explain the mechanisms of six popular algorithms: three discriminative (convolutional neural network (CNN), recurrent neural network (RNN), and Transformer), two generative (variational autoencoder (VAE) and generative adversarial networks (GAN)), and one RL (DQN). I discuss what marketing problems DL is useful for and what fueled its growth in recent years. I emphasize the power and flexibility of DL for modeling unstructured data when formal theories and knowledge are absent. I also describe future research directions.

Construction is a highly dynamic environment with numerous interacting factors that affect construction processes and decisions. Uncertainty is inherent in most aspects of construction engineering and management, and traditionally, it has been treated as a random phenomenon. However, there are many types of uncertainty that are not naturally modelled by probability theory, such as subjectivity, ambiguity and vagueness. Fuzzy logic provides an approach for handling such uncertainties. However, fuzzy logic alone has some limitations, including its inability to learn from data and its extensive reliance on expert knowledge. To address these limitations, fuzzy logic has been combined with other techniques to create fuzzy hybrid techniques, which have helped solve complex problems in construction. In this chapter, a background on fuzzy logic in the context of construction engineering and management applications is presented. The chapter provides an introduction to uncertainty in construction and illustrates how fuzzy logic can improve construction modelling and decision-making. The role of fuzzy logic in representing uncertainty is contrasted with that of probability theory. Introductory material is presented on key definitions, properties and methods of fuzzy logic, including the definition and representation of fuzzy sets and membership functions, basic operations on fuzzy sets, fuzzy relations and compositions, defuzzification methods, entropy for fuzzy sets, fuzzy numbers, methods for the specification of membership functions and fuzzy rule-based systems. Finally, a discussion on the need for fuzzy hybrid modelling in construction applications is presented, and future research directions are proposed.