Transforming the Knowledge of Smart Sustainable Cities

CONCEPTUAL AND THEORETICAL BACKGROUND

A new techno-urban phenomenon dubbed “smart cities sustainability” has emerged since the mid-2010s, bringing together sustainability, urbanization, and information and communication technology (ICT). When they are put together as set framework, they combine the best aspects of sustainable cities with the innovative ICT solutions produced in cities. More and more research institutions and colleges are focusing on creating innovative, sustainable cities to solve sustainability and urbanization’s impending issues.

Research question

How could we support to the objective of sustainable growth and gain a rational data-driven understanding of sustainability issues?

Analytics of Big Data

“Big data” refers to datasets that are both too huge and complicated to be processed by traditional data processing methods in their current form. This suggests that current computer paradigms and procedures are insufficient for dealing with massive amounts of information. There are a number of Vs that may be used to describe big data, the most prominent of which are volume, variety, and speed.

Data mining, machine learning, statistical analysis, regression analysis, and data warehousing are some of the data science foundations utilized in urban analytics. These approaches can be used alone or in combination. The application of these methods is dependent on the urban context and the specifics of the problem being addressed. In contrast to other forms of analytics, data mining concentrates on the automated discovery and extraction of relevant information from large datasets. Data mining, both predictive and descriptive, is a significant focus of this research.

PROBLEMS OF URBAN SUSTAINABILITY TO DATA MINING TASKS

The issues of urban sustainability pertaining to varied urban realms and their connectivity or coordination are supported by typical data mining jobs. Data scientists can decompose an urban sustainability difficulty into subtasks, from which the solution can be assembled to resolve the issue (Bibri & Krogstie, 2018). Energy, transportation, mobility, infrastructure, education, the built environment, healthcare, and public safety are all potential topics of discussion. What sets data scientists apart is their ability to break down a data analysis challenge into sub-tasks. Available tools and approaches can be employed either alone or in combination. Several of these sub-jobs are still conventional data mining tasks, while others are unique to the specific setting of an urban sustainability challenge.

A methodology for data mining in urban analytics

This section explains the core principles of data science that underlie everyday data mining activities. An urban analytics framework (see Figure 1) is then offered, which organizes the issues surrounding urban sustainability to promote replicability, reproducibility, and objectivity.

Figure 1: A methodology for data mining in urban analytics

Numerous codifications of data mining define the process as a set of well-defined processes, which include problem comprehension and preparation, understanding of data and preparation, creation of model, evaluation of results, as well as deployment. By strengthening intelligent decision support, data mining intends to enhance the control, effectiveness, administration, and urban systems planning, the environment and human services connected to water, education, healthcare, energy, and safety, among others.

Data Mining and its Methods

An essential part of data mining is finding patterns in vast datasets that are otherwise unnoticed and then using this information in decision-making processes by constructing meaningful connections and presenting the findings in a way that is easy to understand. Numerous data mining techniques exist for the analysis and processing of massive volumes of data, notably grid mining, distributed multi-layer data mining, and data mining, (Bibri & Krogstie, 2018). It is possible to employ a range of data mining algorithms to solve challenges related to urban sustainability. These include classification and clustering as well as regression and profiling.

Descriptive and prescriptive data mining examples for smart sustainable cities are provided below.

Predictive questions:

• Identify and classify the various modes of transportation used by the general public.
• Classify the use of electricity in the home.
• Forecast GHG emissions over the next month or year.
• Forecast traffic congestion in the immediate future.

Descriptive questions:

• Find helpful travel behavior categories.
• Look for unusual patterns of collective or individual movement.
• Find out what causes gridlock and how to avoid it.
• In metropolitan regions with a mix of land use, describe the normal accessibility of facilities.
• Find groupings of people who travel in similar ways.
• Discover the distinct subsets of travel based on their behavior, duration, and goal.

Evaluating and Interpreting the Obtained Results

The quick use of data mining results for decision-making purposes is not recommended. Results must be evaluated prior to deployment to verify that the developed models fulfil the desired urban goals in respect of decision-making assistance. The aim is to seek an effective urban sustainability dilemma solution of data-analytics. When it pertains to urban sustainability, data mining is typically only a small component of a larger solution. And it needs to be examined in this way. Various extrinsic factors may make them ineffective when analyzing models, even if they pass rigorous laboratory evaluations.

Results Application to Urban Operations, Strategies, Functions, Services, and Policies

Models for predicting and describing urban life, or information systems for public services, are the most prominent examples of how this technology can be used to support decision-making in various urban entities. The design and management of public transportation systems could benefit from using a model that predicts or describes travel behavior. Data on travel patterns could be beneficial in identifying potential disruptions to the transportation system (Bibri & Krogstie, 2018). It’s essential to think creatively about how people navigate the various transportation systems, just as it’s essential to assign passengers on different lines to measure the number of passengers on each line accurately. We still have a long way to go in understanding complex-environment navigation.

CONCLUSION

Smart sustainable cities of the coming will rely significantly on data-driven decision-making processes, systems, and procedures. Our understanding of these technologies is at a critical crossroads. With the use of big data analytics as well as data mining methods that use and combine multiple jobs to handle, smart sustainable towns will be better managed, managed, planned, created and controlled. They deal with a variety of urban sustainability-related issues.

References

Bibri, S. E., & Krogstie, J. (2018, October). The big data deluge for transforming the knowledge of smart sustainable cities: A data mining framework for urban analytics. In Proceedings of the 3rd International Conference on Smart City Applications (pp. 1-10).