KPMG (Netherlands)

The EU Green Deal mirrors the societal concern regarding climate change. In the software industry this means reducing the energy consumption of a program. This can be done by looking both at the hardware and the software used. We take a software approach and focus on the programming language choices and how a computer program is written. The programming languages we investigate are Java, JavaScript, Python, PHP, Ruby, C, C++ and C#. We compare programs from The Computer Language Benchmarks Game and these programs are divided in groups that solve the same problem.We find that the choice of programming language and the way of writing your program influences the energy consumption. The programming languages that consume the least amount of energy over all the problems are C and C++, albeit these programming languages consume more energy when compilation flags are minimized. However, in some cases we could not prove that there was a difference in the energy consumption. We also find that the hardware used to run the programs has an influence on the energy consumption. We find that the runtime of a program correlates with the energy consumption, but a shorter runtime does not necessarily result in a lower energy consumption.

Inpatient violence is a common and severe problem within psychiatry. Knowing who might become violent can influence staffing levels and mitigate severity. Predictive machine learning models can assess each patient’s likelihood of becoming violent based on clinical notes. Yet, while machine learning models benefit from having more data, data availability is limited as hospitals typically do not share their data for privacy preservation. Federated Learning (FL) can overcome the problem of data limitation by training models in a decentralised manner, without disclosing data between collaborators. However, although several FL approaches exist, none of these train Natural Language Processing models on clinical notes. In this work, we investigate the application of Federated Learning to clinical Natural Language Processing, applied to the task of Violence Risk Assessment by simulating a cross-institutional psychiatric setting. We train and compare four models: two local models, a federated model and a data-centralised model. Our results indicate that the federated model outperforms the local models and has similar performance as the data-centralised model. These findings suggest that Federated Learning can be used successfully in a cross-institutional setting and is a step towards new applications of Federated Learning based on clinical notes.

Abstract An interesting relatively new development in the field of corporate climate change disclosures is the Task force on Climate-related Financial Disclosures (TCFD). The TCFD aims to help identify the information needed by financial stakeholders to appropriately assess and price climate change related risks and opportunities. In its first Report (2016), the TCFD recommends that companies provide climate change related disclosures specifying the impact thereof on their financial performance through mainstream (i. e. public) financial filings. In this paper, we look at the financial accounting standards as an institutional framework, and in particular pose the question to what extent this framework supports companies to disclose how climate change impacts their operations and the value of the production assets. To test to what extent companies make disclosures in relation to climate change, we selected four energy companies and conducted a comparative case study analysis. Our focus is on the valuation of production assets, more specifically, drilling platforms, windmill platforms, heavy equipment and transport means used to support the production, and pipes and cables to transport the energy units produced. Interesting findings were: (i) in all four cases, potential future changes (caused by climate change) concerning the valuation of the production assets are not (yet) accounted for in their Balance Sheet Annex. This is remarkable because climate change is likely to have an effect on the future value of the production assets employed in the two types of industries, among others caused by the development that renewable energy demand increases at the expense of non-renewable energy demand; and (ii) the current financial reporting system does not support renewable energy companies to provide meaningful and quantitative insights in expected increases of their future cash inflows and their financial and innovation potential. This impedes financiers and investors to accurately and meaningfully assess the value of a renewable energy company’s business compared with a non-renewable company’s business.

We address the problem of detecting highly raised crowd density in situations such as indoor dance events. We propose a new method for estimating crowd density by anonymous, non-participatory, indoor Wi-Fi localization of smart phones. Using a probabilistic model inspired by statistical mechanics, and relying only on big data analytics, we tackle three challenges: (1) the ambiguity of Wi-Fi based indoor positioning, which appears regardless of whether the latter is performed with machine learning or with optimization, (2) the MAC address randomization when a device is not connected, and (3) the volatility of packet interarrival times. The main result is that our estimation becomes more—rather than less—accurate when the crowd size increases. This property is crucial for detecting dangerous crowd density.

Specific cyber risks can and must be treated by applying the necessary measures and by reacting effectively when an organization is subjected to cyber attack. This can be done by focusing cybersecurity efforts to protect the most crucial and sensitive assets-an organization's crown jewels . Besides being able to respond to incidents, it is crucial for organizations to remain up-to-date and informed of emerging threats, and to learn from other organizations how best to react to incidents. The starting point of managing cyber risks is the same starting point of managing enterprise, defining the organization's risk appetite, an exercise to be performed by business representatives instead of cybersecurity specialists.

Life scientists and bioinformaticians increasingly rely on web-based resources. The number of such resources, as well as the amount of available content, grows continuously. The need for efficient interoperability is becoming important. Moreover, as data propagate easily through such media, their quality and pertinence need to be monitored closely. It costs money to set up and maintain a resource. Because biology, like any other field of scientific inquiry, is very dynamic, maintenance is labour-intensive. Users of a resource will generally want to be assured of its quality; in other words, a scheme of quality assurance has to be in place. Governance, or who is responsible for what, has to be clear for the community. A business model addresses these issues in their mutual dependence. Therefore, every resource comes with a business model, irrespective of whether or not its creators are aware of it.

Studies adoptions of four different payment systems in ten countries during 1988‐96. Adoption curves are estimated from secondary data concerning ATMs and banker’s cards. Usage of a previously introduced payment system appeared to affect the adoption of a system innovation. Based on the shape of the adoption curves and the estimation of external and internal influences from the social system, it was concluded that the adoption process was almost exclusively driven by internal influences, i.e. social learning. Draws conclusions for the introduction of the smart card.

Despite the trend towards an increasing use of strategic technology alliances, mortality rates of cooperative agreements have always been extremely high. In this paper we argue that rapid economic and technological developments have overthrown traditional thinking about alliances and that firms could benefit from a new perspective on partnering. This new approach is referred to as High Touch Partnering. The new framework stresses the need for a balanced attention to strategic (re‐)positioning, establishing adequate alliance capabilities, building business communities with partners and improved partner selection.

Based on the results from an ethics audit, organizations can take concrete measures and develop activities to improve the ethical content. In this chapter, a workable ethics process will be discussed. The ethics process (or ethics project or ethics program) is a cluster of measures and activities used to protect and improve the ethics of an organization. The measures and activities that can be implemented will come up for discussion in the following chapter. During the ethical development process, a number of difficult choices have to be faced. In this chapter, I describe these choices as areas of tension which should be “relieved” during an ethics process. Ignoring one or more of these conflicting issues or making a choice that has not been thought through has consequences for the effectiveness of the process. In this chapter, a vision of the management of ethics will also be discussed. This vision provides guidelines for considering options in regards to the conflicting issues which have been outlined. This chapter closes with a description of a number of simple ethics project steps that have been taken in several organizations. In the introductory chapter of this book it was already stated that hardly any explicit or implicit attention is paid in business literature to the assumptions with regard to the management of ethics and the problems which can occur in this field.