1,721,087 research outputs found
Characterizing the technological evolution of smartphones
No full textRecent technological advancements in smartphone have paved the way for the rapidly growing mobile commerce. As smartphone vendors launch the products with a rich variety of technical features for different end-user market segments, understanding the evolution of these features is of vital importance to all stakeholders in the smartphone industry. We address this issue by exploring technical specifications of smartphones at both the feature and the device level. In particular, we introduce the benchmarks to operationalize the overall performance of smartphone models, such that multidimensional technical features can be quantitatively summarized into a single index. Through the analysis of a comprehensive dataset entailing technical features for smartphone models launched during the years 2012-2015, we show that although certain features have become the standard functionality, the smartphone industry is largely innovative and continues to evolve over time. We believe our findings may provide important insights into the future development and design strategies of smartphones
Which kind of legal scholarship do judges cite? A data-driven analysis of Portuguese high civil court decisions
Full text linkThe Portuguese judicial system is based on the German juridical system and is one of the only
jurisdictions where it is common for judges to cite academic works. Given the importance of a non biased juridical system, this paper aims to examine the Supreme Court rulings documented in the
legal database to extensively analyze these documents to study author citations while also shedding
light on the use of legal scholarship by judges. This database collects over 60000 legal case
decisions solved by the Portuguese Supreme Court (“Supremo Tribunal de Justiça”). It enables us
to identify the authors referenced in court, check the connections between the citations and the
judges, and draw inferences using diverse Data Analytics and Network Graph tools and techniques.
Keywords: Portuguese Supreme Court, Most Cited Authors, Network Graph, Page Rank
Classification
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Incorporating complex domain knowledge into a recommender system in the healthcare sector
Full text linkIn contrast to other domains, recommender systems in health sector may benefit particularly from the incorporation of medical domain knowledge, as it provides meaningful and personalised recommendations. With recent advances in the area of representation learning enabling the hierarchical embedding of health knowledge into the hyperbolic Poincaré space, this thesis proposes a recommender system for patient-doctor matchmaking based on patients’ individual health profiles and consultation history. In doing so, a dataset from a private healthcare provider is enriched with Poincaré embeddings of the ICD-9 codes. The proposed model outperforms its conventional counterpart in terms of recommendation accuracy
Quantifying Covid-19 impact on Airbnb hosting: Lisbon as a case study
Full text linkWhile Covid-19 impact on tourism and the sharing economy has proven to be significant by plenty of previous research, data and tools to recursively measure financial impact are missing in the current state of knowledge. This paper aims at quantifying the disease’s financial impact on Airbnb prices, bookings and hosting revenues with machine learning. The bottom-up approach used predicts a city’s losses at listing level over time and therefore grants leeway to analyzing impact across various dimensions. The city of Lisbon is used to showcase the model’s performance and versatility of results
Data-driven modeling of smart builiding energy management
Full text linkBuildings consume approximately 40% of energy in total, which contributes negatively to the environment. Building Energy Management Systems(BEMS) have been used to monitor energy consumption and increase usage efficiency. In this study, the components and importance of BEMS are emphasized. The data from the management systemoftheChamchuri5building in Chula long korn University, Thailand, were used as a template for data-driven modeling for energy usage in smart buildings to analyze the patterns of energy consumption. Using multilevel modeling on theChamchuri5 building ,the main factors that consume energy on a macro and micro level are analyzed .Energy variation between zones and floors was spotted
An analysis of the impact of ESG on corporate financial performance in Europe
Full text linkESG investing has experienced extensive growth in the last decade. This paper analyses the impact of ESG on CFP for the years 2015-2020, using a dataset by Refinitiv with5676observationsin the EU. Using a comprehensive dataset, this paper can confirm that ESG still has a positive impact on CFP, despite its extensive growth and increasing implementation. Furthermore, by analyzing the sub categories of ESG, this paper shows which factors contribute to this positive relationship. Also, the inclusion of COVID-19dataoffersunique possibilities to contribute to the existing literature, as the wealth-protective impact of ESG Scores can be evaluated
Demand shaping in practice - investigating the use of predictive models to identify causal relationships
No full textVOIDS provides deep learning-based demand forecasting. To provide their customers with
countermeasures in response to different supply/demand scenarios, VOIDS needs to infer the
causal relationship of their clients’ data. This thesis seeks to investigate whether traditional
econometric models as well as newer machine learning models can be used to provide VOIDS
with a scalable solution for doing causal inference for their clients. The joint part will be focused
on theoretical discussions and testing, while individual part 1 compares the results with a
prediction model
A synergistic enhancement to demand forecasting using neural networks with voids - exploring the effects of cross-client data integration
No full textThis paper presents a collaborative effort encompassing four key individual contributions:
optimising feature selection in Temporal Fusion Transformers, enhancing anomaly de tection during special events, examining the impact of Cross-Client data integration on
forecasting accuracy, and leveraging Generative AI for strategic business recommenda tions. Collectively, these studies reveal significant advancements in demand forecasting
and management for e-commerce companies. The results demonstrate improved predic tive accuracy, efficient anomaly handling during critical sales periods, insights into the
benefits and limitations of aggregated data models, and advantages of using generative AI
for recommending business action to mitigate operational risks
- …
