| Big data is the collection of technology generated to store,analyze and handle this bulk data;a macro-tool designed to identify patterns in the chaos of this information flood to develop smart solutions.It is used today in fields as broad as medicine,agriculture,tourism,and protection of the environment.Social media big data make up a huge proportion of the content found on the Internet.Users spend most of their time online nowadays,communicating with them.In Location-based Social Networks LBSNs,users socialize with each other by sharing their current location“geolocation”(also referred to as “check-in”).Understanding this human activity behavior in space and time on LBSN datasets also called “activities behavior” can archive the day-to-day activity patterns,usage behaviors towards social media and presents spatiotemporal evidence of user's daily routines.LBSN also has great importance in environmental studies especially in public green spaces because green parks are important public spaces and play a main role in urban living.The initial point for questioning potential benefits or park utilization in urban green spaces should begin with the evaluation of geographical accessibility of green parks.Traditional strategies for studying the attractiveness of green parks,such as questionnaires and in-situ surveys,are usually insignificant in size,time-consuming and costly,with fewer transferable findings and only site-specific results.This research provides an investigative study by using Location-based social network(LBSN)data to collect spatial and temporal patterns of park visits in ten districts of Shanghai metropolitan city.Throughout the period from July 2014 to June 2017 in Shanghai,China,we investigated the spatiotemporal park visitor's behavior for 122 green parks.Using temporal and spatial analysis of LBSN data,we conduct experimental research on the effects of green spaces on public behavior.The primary objective of this thesis is to observe the user's behavior towards different green spaces in space and time and analyze the spatiotemporal trends,and the distribution changes of check-ins at the city and district level by using LBSN data.We applied linear regression model for the data significance and Kernel density estimation(KDE)is employed to discover the spatiotemporal distribution of check-ins.The results of this research show that LBSN can be considered as a consistent source of information to observe user's spatiotemporal behavior inside a city in space and time.Furthermore,it shows that residents like to visit green parks situated in the city center or downtown area and female users are more likely to visit green parks as compared to male.However,the visitation behavior patterns for male users are quite different during weekdays and weekend as compared to female users,interesting variations in user behavior based on seasonal effects were observed and the study also proved that in2016-2017 visitor's interest increased towards visiting green parks comparing to other two years.In summary,the results of KDE at the district level can better reveal the change trajectory of spatiotemporal patterns of visits at the whole city level.The broad spatial framework of this study provides useful information that could improve urban green space planning and development in other large cities.Through the effective utilization of urban green areas,there is a significant potential to enhance moderate-to-strong physical activity among urban residents.These results illustrate the value of using urban green parks to improve the wellbeing of people who are living in cities around the world. |
PDF Full Text Request