Mobility ChatBot: supporting decision making in mobility data with chatbots

Mobile location data are a major form of Big Data that hold many possibilities for study and insight into human behaviour. This research used mobile location data to investigate the differences in the activity patterns of tourists in Maui, Hawai’i. Mobile data used in this study were app-based location data collected as a stream of mobile phone locations with a timestamp. Tourists were clustered using K-Means based on time spent at attraction types. Different travel experiences were analyzed based on traveler’s accommodation choices, the average distance travelled from accommodation to attraction, and vacation length, which all varied significantly between the tourist clusters. This work provided a new use for K-means clustering with mobile location data to provide insightful information to marketing professionals and tourism management bodies.

Geophysical and remote sensing products that rely on Poisson‐distributed measurement signals, such as cosmic‐ray neutron sensing (CRNS) and gamma spectrometry, often face challenges due to inherent Poisson noise. Common techniques to enhance signal stability include data aggregation or smoothing (e.g., moving averages and interpolation). However, these methods typically reduce the ability to resolve detailed temporal (stationary data) and spatial (mobile data) features. In this study, we introduced a method for contextual noise suppression tailored to Poisson‐distributed data, utilizing a discrete score attribution system. This score filter evaluates each observation against eight different criteria to assess its consistency with surrounding values, assigning a score between 0 (very unlikely) and 8 (very likely) to indicate whether the observation is likely to act as noise. These scores can then be used to flag or remove data points based on user‐defined thresholds. We tested the score filter's effectiveness on both stationary and mobile CRNS data, as well as on gamma‐ray spectrometry and electromagnetic induction (EMI) recordings. In our examples, the score filter consistently outperformed established filters, for example Savitzky–Golay and Kalman, in direct competition when applied to CRNS time series data. Additionally, the score filter substantially reduced Poisson noise in mobile CRNS, gamma‐ray spectrometry and EMI data. The scoring system also provides a context‐sensitive evaluation of individual observations or aggregates, assessing their conformity within the dataset. Given its general applicability, customizable criteria and very low computational demands, the proposed filter is easy to implement and holds promise as a valuable tool for denoising geophysical data and applications in other fields.

Background: Smartphone users frequently connect to the Internet via mobile data or Wi-Fi. Over the past two decades, the worldwide percentage of people who connect to the Internet using their mobile phones has increased drastically.Objective: This study aimed to evaluate the potential link between mobile cellular data/ and Wi-Fi use and adverse health effects.Material and Methods: This cross-sectional study was conducted on 2,796 employees (52% female and 48% male) of Shiraz University of Medical Sciences (SUMS), Shiraz, Iran. The sociodemographic data (e.g., gender, age, nationality, and education level) were collected for all the participants. They were also requested to provide information about their smartphone use including the characteristics of the connection to the Internet using their smartphones (mobile data and Wi-Fi). In addition, the participants’ history of diabetes, hypertension, cardiac ischemia, myocardial infarction, renal failure, fatty liver, hepatitis, chronic lung disease, thyroid disease, kidney stone, gall bladder stone, rheumatoid disease, epilepsy, and chronic headache was recorded through face-to-face interviews.Results: 94% of people participating in this study reported using mobile/Wi-Fi internet. The mean (±SD) Internet usage per day was 117.85±122.70 minutes including 76±98 minutes of mobile data and 42±81 minutes of Wi-Fi use. Conclusion: Our findings showed no link between mobile phone Internet usage and the risk of the above-mentioned health problems. As in 2021, the global average daily time spent on the Internet using mobile phones was 155 minutes, the participants’ lower use time could explain the failure to show any detrimental effects. Considering the study limitations, further large-scale studies are warranted.

As the most immediately impacted population, survivors’ mobility and distribution characteristics are closely linked to earthquake emergency response. Accurate population distribution and mobility data are vital foundational resources for post-disaster decision-making. With the help of mobile phone location data within the earthquake-stricken area, we explore novel rapid assessment approaches to identifying perceived impact area.The widespread adoption of smart mobile devices in China has led to the installation of numerous third-party applications. These applications rely on push notification services enabled by Push Software Development Kits (SDKs) provided by mobile service providers. These SDKs, compliant with data security standards, collection scopes, and transmission protocols, utilize built-in functional modules to periodically collect user-authorized geolocation data. This data encompasses device identifiers, GPS coordinates, WiFi signatures, cellular base station logs, and network connectivity metadata. After encryption, these multi-source data streams inputs are aggregated into structured mobile location datasets. They can provide high-resolution insights into population mobility patterns during disasters.At 9:05 (Beijing time) on 7 January 2025, an earthquake with magnitude Ms 6.8 hit Dingri County, Tibet Autonomous Region, which killed 126 people. We collected a mobile phone location dataset covering a 150 km radius from the epicenter, spanning 38 hours from 9:00 on 6 January to 1:00 on 8 January. It comprises 146,123 records. Each record includes four mobile location-based indicators, namely (1) Active Base Stations (base stations scanned and periodically reported by mobile devices), (2) Active Wi-Fi Hotspots (Wi-Fi hotspots scanned and periodically reported by mobile devices), (3) Mobile Devices (mobile devices that obtain services through multiple positioning methods) and (4) Wi-Fi-Connected Devices (mobile devices connected to Wi-Fi hotspots). Utilizing natural neighbor interpolation and Thiessen polygon interpolation methods, we analyzes changes in four mobile location-based indicators and their spatiotemporal distribution characteristics before and after the earthquake, summarizing crowd movement patterns and communication behaviors after the Dingri earthquake. The results indicate an uneven distribution of population and differing dynamics in mobile phone signal activity. This reflects different behavioral patterns and the potential perceived extent of the earthquake. Within 50 km of the epicenter, all four indicators showed varying degrees of decline post-earthquake, while areas beyond 100 km exhibited short-term surges, reflecting differentiated behavioral responses based on seismic impact severity. In areas experiencing strong shaking, risk avoidance behavior predominated, while in areas where shaking was noticeable but less severe, communication behavior was more prominent. Mobile data decline zones showed high spatial correlation with intensity VIII+ regions, proving their effectiveness as rapid indicators for identifying strongly affected areas. Notably, mobile location data enabled accurate identification of strongly affected zones within 30 min post-earthquake.The research establishes a theoretical-technical framework supporting three critical post-disaster applications: (1) dynamic population distribution sensing, (2) behavioral pattern analysis of affected populations, and (3) rapid evaluation of seismic perception zones.

To evaluate the association between social distancing quantified by mobile phone data and the current prevalence of COVID-19 infections in the U.S. per capita. Data were accessed on April 4, 2020, from Centers for Disease Control and Prevention, Google COVID-19 Community Mobility Report, and the United States Census Bureau to report prevalence of COVID-19 infections, mobility data, and population per state, respectively. Mobility data points were defined as daily length of visit or time spent in a single location based on mobile phone users shared locations from February 7 - March 29, 2020. Multivariable linear regression was used to evaluate relationships between normalized per capita infection prevalence and six parameters of social distancing. Mobility data indicated the following percent changes compared to median values of baseline activity: -50% in transit stations, -45% in retail/recreation, -36% in workplaces, -23% in grocery/pharmacy, -19% in parks, and +12% in residential living areas. Multivariable linear regression revealed significant correlation between prevalence of infection per capita and parameters of social distancing (R= 0.604, P= 0.002). Time at home was not an independent predictor for prevalence of infection per capita (beta= 0.016; 95% CI, -0.003 to 0.036; P= 0.09). Based on mobility reports from mobile phone GPS data and six characteristics of social distancing, significant associations were identified between geographic activity and prevalence of COVID-19 infections in the U.S. per capita. Mobile phone data utilizing 'location history' may be warranted to monitor the effectiveness of social distancing parameters on reducing prevalence of COVID-19 in the U.S.

Human Mobility Data in the COVID-19 Pandemic: Characteristics, Applications, and Challenges

Traditionally, the equipment used to measure air pollution is expensive and placed around cities or in mobile laboratories. It might only represent a certain area and not the entire city due to the locations and limited number of monitoring stations. Nowadays, a mobile sensing is becoming an alternative option to monitor air quality in urban environment due to its ease of use, high flexibility, and low price. This paper develops a vehicular-based mobile monitoring system for real time air quality sensing and visualization across large cities with high spatial resolution. The commercially available low-cost CO, NO[Formula: see text], NH[Formula: see text] O[Formula: see text], CH[Formula: see text], SO[Formula: see text], PM[Formula: see text], temperature and humidity sensors along with the microcontroller and GPS were integrated in a sensing device installed on the roof of taxi and sport utility vehicle (SUV). The developed device was calibrated through a reference monitoring station and validated through field measurement. We first split the entire city with a uniform grid discretization. We then propose a data processing methodology based on machine learning algorithms for generating 250 representative data set from 286 million data which is collected using the vehicular based mobile sensors. Next we present the representativeness of the data set by comparison of stationary data and mobile data. We also describe the analytical results and spatial distribution with high spatial resolution throughout the city. In addition, the collected mobile sensor data is also used to show that the significant differences and spatial variability in mean levels per street. Finally, we conclude that the proposed mobile monitoring system using high spatial resolution can effectively map the air quality in metropolitan environment and provide detail about the spatial variability that cannot be done with stationary monitoring systems.

The emergence of SARS-CoV-2 into a highly susceptible global population was primarily driven by human mobility-induced introduction events. Especially in the early stages, understanding mobility was vital to mitigating the pandemic prior to widespread vaccine availability. We conducted a systematic review of studies published from January 1, 2020, to May 9, 2021, that used population-level human mobility data to understand SARS-CoV-2 transmission. Of the 5505 papers with abstracts screened, 232 were included in the analysis. These papers focused on a range of specific questions but were dominated by analyses focusing on the USA and China. The majority included mobile phone data, followed by Google Community Mobility Reports, and few included any adjustments to account for potential biases in population sampling processes. There was no clear relationship between methods used to integrate mobility and SARS-CoV-2 data and goals of analysis. When considering papers focused only on the estimation of the effective reproductive number within the US, there was no clear relationship identified between this measure and changes in mobility patterns. Our findings underscore the need for standardized, systematic ways to identify the source of mobility data, select an appropriate approach to using it in analysis, and reporting.

Given the rapid cross-country spread of SARS-CoV-2 and the resulting difficulty in tracking lineage spread, we investigated the potential of combining mobile service data and fine-granular metadata (such as postal codes and genomic data) to advance integrated genomic surveillance of the pandemic in the federal state of Thuringia, Germany. We sequenced over 6500 SARS-CoV-2 Alpha genomes (B.1.1.7) across 7 months within Thuringia while collecting patients’ isolation dates and postal codes. Our dataset is complemented by over 66,000 publicly available German Alpha genomes and mobile service data for Thuringia. We identified the existence and spread of nine persistent mutation variants within the Alpha lineage, seven of which formed separate phylogenetic clusters with different spreading patterns in Thuringia. The remaining two are subclusters. Mobile service data can indicate these clusters’ spread and highlight a potential sampling bias, especially of low-prevalence variants. Thereby, mobile service data can be used either retrospectively to assess surveillance coverage and efficiency from already collected data or to actively guide part of a surveillance sampling process to districts where these variants are expected to emerge. The latter concept was successfully implemented as a proof-of-concept for a mobility-guided sampling strategy in response to the surveillance of Omicron sublineage BQ.1.1. The combination of mobile service data and SARS-CoV-2 surveillance by genome sequencing is a valuable tool for more targeted and responsive surveillance.

The mobile network traffic patterns in urban areas significantly diverge depending on commercial and residential establishments. These regional traffic patterns provide crucial clues for predicting traffic patterns precisely. Previous studies have employed a combination of time-series and convolutional Deep Learning (DL) models to effectively capture the correlation of the regional features and traffic patterns. Despite promising results, these approaches are limited in identifying pattern similarities among sparsely located regions and can be further improved. To this end, this study proposes a GEospatial clustering and residual COnvolutional temporal long Short-term memory (GECOS) framework consisting of clustering and DL components. The proposed Urbanflow Peak Clustering (UPC) component exploits the peak traffic times of daily mobile data to obtain the groups of cells with similar traffic patterns apart from their geographical diversity. The UPC improves the scalability of existing algorithms and enables DL components to improve their accuracy by recognizing unique regional patterns and localizing the training targets. The proposed Residual Convolutional TCN-LSTM (RCTL) serves as the DL component of GECOS that improves TCN-LSTM structure through layer-wise feature transfer and enhances long-term dependency learnability. The RCTL ensures more accurate capturing of extensive spatiotemporal features through structural enhancements. The experiments conducted on real-world mobile traffic data showcase 43% improvement by GECOS compared to state-of-the-art models, enabling precise traffic engineering policies by operators.

Researchers face the trade-off between publishing mobility data along with their papers while protecting the privacy of the individuals. In addition to the anonymization process, other techniques, such as spatial discretization and location concealing or removal, are applied to achieve these dual objectives. The primary research question is whether concealing the observation area is an adequate form of protection or whether human mobility patterns in urban areas are inherently revealing of location. The characteristics of the mobility data, such as the number of activity records in a given spatial unit, can reveal the silhouette of the urban landscape, which can be used to infer the identity of the city in question. The presented locating method was tested on multiple cities using different open datasets and coarser spatial discretization units. While publishing mobility data is essential for research, concealing the observation area is insufficient to prevent the identification of the urban area. Instead of obscuring the observation area, noise should be added to the trajectories to mitigate privacy risks regarding the individuals.

Narrowing the 'Digital Divide': The Role of Complementarities between Fixed and Mobile Data in South Africa

We explore relationships between health information seeking activities and engagement with healthcare professionals via a privacy-sensitive analysis of geo-tagged data from mobile devices. We analyze logs of mobile interaction data stripped of individually identifiable information and location data. The data analyzed consist of time-stamped search queries and distances to medical care centers. We examine search activity that precedes the observation of salient evidence of healthcare utilization (EHU) (ie, data suggesting that the searcher is using healthcare resources), in our case taken as queries occurring at or near medical facilities. We show that the time between symptom searches and observation of salient evidence of seeking healthcare utilization depends on the acuity of symptoms. We construct statistical models that make predictions of forthcoming EHU based on observations about the current search session, prior medical search activities, and prior EHU. The predictive accuracy of the models varies (65%-90%) depending on the features used and the timeframe of the analysis, which we explore via a sensitivity analysis. We provide a privacy-sensitive analysis that can be used to generate insights about the pursuit of health information and healthcare. The findings demonstrate how large-scale studies of mobile devices can provide insights on how concerns about symptomatology lead to the pursuit of professional care. We present new methods for the analysis of mobile logs and describe a study that provides evidence about how people transition from mobile searches on symptoms and diseases to the pursuit of healthcare in the world.

In large-scale wireless sensor networks the data- gathering mechanism to introducing mobility into the network. . We consider the location service in a WSN, where each sensor needs to maintain its location information by 1) frequently updating its location information within its neighboring set of polling points in the network. In the previous work a mobile data collector, for convenience called an M-collector for gathering information from the sensor and send to sink node. M- collector could be a mobile robot or a vehicle equipped with a powerful transceiver and battery, working like a mobile base station and gathering data while moving through the field. An M-collector starts the data-gathering tour periodically from the static data sink, polls each sensor while traversing its transmission range, then directly collects data from the sensor in single-hop communications, and finally transports the data to the static sink. Since data packets are directly gathered without relays and collisions, the lifetime of sensors is expected to be prolonged. In this paper, we mainly focus on the problem of minimizing the length of each data-gathering tour and remove the redundancy for data gathering. We develop a stochastic sequential decision framework to analyze this problem. Under a Markovian mobility model, the location update decision problem is modeled as a Markov Decision Process (MDP). For the work with strict distance/ time and energy constraints, we consider utilizing single and multiple M-collectors and propose a data- gathering algorithm where multiple M-collectors traverse through several shorter subtours concurrently to satisfy the distance/time and energy constraints. To identify the redundancy of data from the sensor using Pearson auto correlated technique. Our proposed system for mobile data gathering scheme can improve the scalability and balance the energy consumption among sensors.

Genomics, combined with population mobility data, used to map importation and spatial spread of SARS-CoV-2 in high-income countries has enabled the implementation of local control measures. Here, to track the spread of SARS-CoV-2 lineages in Bangladesh at the national level, we analysed outbreak trajectory and variant emergence using genomics, Facebook ‘Data for Good’ and data from three mobile phone operators. We sequenced the complete genomes of 67 SARS-CoV-2 samples (collected by the IEDCR in Bangladesh between March and July 2020) and combined these data with 324 publicly available Global Initiative on Sharing All Influenza Data (GISAID) SARS-CoV-2 genomes from Bangladesh at that time. We found that most (85%) of the sequenced isolates were Pango lineage B.1.1.25 (58%), B.1.1 (19%) or B.1.36 (8%) in early-mid 2020. Bayesian time-scaled phylogenetic analysis predicted that SARS-CoV-2 first emerged during mid-February in Bangladesh, from abroad, with the first case of coronavirus disease 2019 (COVID-19) reported on 8 March 2020. At the end of March 2020, three discrete lineages expanded and spread clonally across Bangladesh. The shifting pattern of viral diversity in Bangladesh, combined with the mobility data, revealed that the mass migration of people from cities to rural areas at the end of March, followed by frequent travel between Dhaka (the capital of Bangladesh) and the rest of the country, disseminated three dominant viral lineages. Further analysis of an additional 85 genomes (November 2020 to April 2021) found that importation of variant of concern Beta (B.1.351) had occurred and that Beta had become dominant in Dhaka. Our interpretation that population mobility out of Dhaka, and travel from urban hotspots to rural areas, disseminated lineages in Bangladesh in the first wave continues to inform government policies to control national case numbers by limiting within-country travel.