Publicação
Uncertainty Visualization using Hypothetical Outcome Plots
| Resumo: | Data-driven decision-making is crucial for any business. With an increasing interest in Business Intelligence, Data Visualization is playing a major role in decision-making processes. For making a well-informed and accurate decision, it is important to understand uncertainty in the data visualizations. Uncertainty visualizations improve the way users understand the data, as well as the confidence in their conclusions. An important type of uncertainty visualizations is the Hypothetical Outcome Plots (HOPs), which allow the audience to gain an intuitive idea of uncertainty through animated sequences of random draws from a distribution, leading to a more accurate understanding and decision. This document intends to detail a proof-of-concept by carrying out a comparison of static visualization vs. HOPs in terms of efficiency and accuracy of results interpretation for Wayne Enterprises (fictional name) forecasting projects, in particularly the ones related with product launches and product loss of exclusivity. Wayne Enterprises is a world-leading supplier of advanced analytics, technological services and clinical investigation solutions for the life sciences industry. For that objective, it was built two prototypes using Python to support the proof-of-concept execution. A between-group experiment was carried out with 40 members of the German consulting team of Wayne Enterprises, where half answered a survey based on static visualizations and the other half based on HOPs. From this experiment, it is possible to conclude that HOPs can achieve similar results that static visualizations, with people taking the decision in less than half of the time when visualizing a HOP. Thus, it is possible to improve Wayne Enterprises decision-making process by accelerating it with Hypothetical Outcome Plots. |
|---|---|
| Autores principais: | Pereira, Ana Patrícia Gonçalves |
| Assunto: | Data Visualization Uncertainty Decision-Making Hypothetical Outcome Plots Python Visualização de dados Incerteza Tomada de decisões |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | Data-driven decision-making is crucial for any business. With an increasing interest in Business Intelligence, Data Visualization is playing a major role in decision-making processes. For making a well-informed and accurate decision, it is important to understand uncertainty in the data visualizations. Uncertainty visualizations improve the way users understand the data, as well as the confidence in their conclusions. An important type of uncertainty visualizations is the Hypothetical Outcome Plots (HOPs), which allow the audience to gain an intuitive idea of uncertainty through animated sequences of random draws from a distribution, leading to a more accurate understanding and decision. This document intends to detail a proof-of-concept by carrying out a comparison of static visualization vs. HOPs in terms of efficiency and accuracy of results interpretation for Wayne Enterprises (fictional name) forecasting projects, in particularly the ones related with product launches and product loss of exclusivity. Wayne Enterprises is a world-leading supplier of advanced analytics, technological services and clinical investigation solutions for the life sciences industry. For that objective, it was built two prototypes using Python to support the proof-of-concept execution. A between-group experiment was carried out with 40 members of the German consulting team of Wayne Enterprises, where half answered a survey based on static visualizations and the other half based on HOPs. From this experiment, it is possible to conclude that HOPs can achieve similar results that static visualizations, with people taking the decision in less than half of the time when visualizing a HOP. Thus, it is possible to improve Wayne Enterprises decision-making process by accelerating it with Hypothetical Outcome Plots. |
|---|