The College Experience Study represents the most extensive longitudinal mobile sensing study to date, leveraging passive and automatic sensing data from the smartphones of over 200 Dartmouth students across five years (2017 - 2022). This groundbreaking research aimed to assess their mental health (e.g., depression, stress), the impact of COVID-19, and general behavioral trends.

The study's importance has been magnified during the global pandemic, necessitating a better understanding of mental health dynamics among undergraduate students throughout their college years. By tracking two cohorts of first-year students both on and off campus, we have accumulated a rich dataset offering insights into changing behaviors, resilience, and mental health in college life. We hope that this dataset will serve as a cornerstone for researchers, educators, and policymakers alike, seeking to enhance their understanding and interventions related to student mental health and behavior.

This dataset is unique for several reasons. It encompasses deep phone sensing data and self-reports spanning four continuous years for 200 undergraduate students at Dartmouth College, both during term time and breaks. Additionally, it incorporates periodic brain-imaging data for this cohort of students, along with surveys. The College Experience dataset enables researchers to explore numerous issues in behavioral sensing and brain imaging to advance our understanding of college students' mental health.

:rocket: Updates

• Apr 15th 2025: Raw app usage (i.e., list of running apps) is now available.
• Dec 5th 2024: Raw call logs, sms logs, and unlocks are now available.
• Oct 27th 2024: Raw sensing data will be released in batches over the next few weeks!

Content

College Experience Study makes use of the StudentLife app, developed for Android and iOS, autonomously capturing a variety of human behaviors 24/7, including:

• Bed time, wake up time, and sleep duration
• The number of conversations and the duration of each conversation per day (Android only)
• Physical activity (walking, sitting, running, standing)
• Locations visited and duration of stay (e.g., dorm, class, party, gym)
• Stress levels over weeks and throughout college
• App usage (Android only)
• COVID concern
• and more

In addition to passive sensing data, our study also involved gathering responses from detailed surveys and conducting brain scans throughout the research period. These diverse data sources can be used together to uncover insightful correlations and draw meaningful conclusions. An illustrative example of this potential is explored in the study "Predicting Brain Functional Connectivity Using Mobile Sensing", which demonstrates how mobile sensing data can predict brain functional connectivity, offering new avenues for understanding mental health conditions.

Data Availability

Note: Some features are exclusive to Android phones. Each folder includes a data definition file detailing the features and their availability across Android and iOS. Also, note that some features like conversation tracking initially covered both user groups but were later restricted due to iOS policy changes so they might be available for iOS users only during the beginning of the study.

For more details, refer to the College Experience Study paper and the original StudentLife website.

Term Definitions and Academic Calendars

For additional context and understanding of the timeline relevant to the dataset, below are the archived links to Dartmouth College's calendars. These archives provide an overview and detailed breakdown of significant dates for each academic year covered by the study:

Context

StudentLife is the first study that uses passive and automatic sensing data from the phones of a class of 48 Dartmouth students over a 10 week term to assess their mental health (e.g., depression, loneliness, stress), academic performance (grades across all their classes, term GPA and cumulative GPA) and behavioral trends (e.g., how stress, sleep, visits to the gym, etc. change in response to college workload -- i.e., assignments, midterms, finals -- as the term progresses).

Much of the stress and strain of student life remains hidden. In reality faculty, student deans, clinicians know little about their students outside of the classroom. Students might know about their own circumstances and patterns but know little about classmates. To shine a light on student life we develop the first of a kind StudentLife smartphone app and sensing system to automatically infer human behavior. Why do some students do better than others? Under similar conditions, why do some individuals excel while others fail? Why do students burnout, drop classes, even drop out of college? What is the impact of stress, mood, workload, sociability, sleep and mental health on academic performance (i.e., GPA)? The study used an android app we developed for smartphones carried by 48 students over a 10 week term to find answers to some of these pressing questions.

Content

The StudentLife app that ran on students' phones automatically measured the following human behaviors 24/7 without any user interaction: - bed time, wake up time and sleep duration - the number of conversations and duration of each conversation per day - physical activity (walking, sitting, running, standing) - where they were located and who long they stayed there (i.e., dorm, class, party, gym) - the number of people around a student through the day - outdoor and indoor (in campus buildings) mobility - stress level through the day, across the week and term - positive affect (how good they felt about themselves) - eating habits (where and when they ate) - app usage - in-situ comments on campus and national events: dimension protest, cancelled classes; Boston bombing.

Other Links

For more detail, please check the original StudentLife Study Website.

Rui Wang, Fanglin Chen, Zhenyu Chen, Tianxing Li, Gabriella Harari, Stefanie Tignor, Xia Zhou, Dror Ben-Zeev, and Andrew T. Campbell. "StudentLife: Assessing Mental Health, Academic Performance and Behavioral Trends of College Students using Smartphones." In Proceedings of the ACM Conference on Ubiquitous Computing. 2014. Nominated for the best paper award (top 5% of all papers) pdf

Data Contact

Andrew Campbell, Dartmouth College Weichen Wang (weichen.wang.gr@dartmouth.edu)

StudentLife Team Dror Ben-Zeev, Andrew Campbell, Fanglin Chen, Zhenyu Chen, Tianxing Li, Rui Wang and Xia Zhou (Dartmouth College), Gabriella Harari (University of Texas at Austin), Stefanie Tignor (Northeastern University)

We would like to thank the following people for their input and guidance is getting the study going. Ethan Berke (DHMC), Tanzeem Choudhury (Cornell), Randy Colvin (Northeasten), Sam Gosling (UT Austin) and Catherine Norris (Swarthmore)

This dataset contains cross-sectional survey data collected from 925 Chinese college students examining the relationships between mobile phone addiction, aggression, self-control, and psychological symptoms. Data were collected in April 2024 through an online survey system accessible via computers, iPads, and smartphones. Participants were recruited from humanities disciplines (Chinese Language and Literature, Radio and Television Studies, and related fields) at a university in western China. The sample comprised students across different educational levels: 668 bachelor's degree students (72.22%), 160 associate-to-bachelor transfer students (17.30%), and 97 master's students (10.49%), all in their first or second years of study. The dataset includes responses to four validated psychological instruments: the 17-item Mobile Phone Addiction Index Scale (MPAIS), the 22-item Buss-Perry Aggression Questionnaire (BPAQ), the 19-item Self-Control Scale (SCS), and the 90-item Symptom Checklist-90 (SCL-90). Additionally, comprehensive demographic information was collected, including age, gender, educational level, family income, parental education levels, and family structure variables. All responses were recorded on 5-point Likert scales with appropriate reverse coding applied where necessary. The tabular dataset contains 925 rows (individual participants) and approximately 170 columns representing survey items, computed subscale scores, total scores, and demographic variables. Missing data analysis revealed less than 5% missing values distributed completely at random, with 25 incomplete responses excluded from the final dataset. Data are stored in SPSS (.sav) format and CSV format for broader accessibility. The dataset enables replication of the reported moderated mediation analyses using PROCESS macro Model 14, as well as network analyses examining symptom-level interactions. All data have been de-identified to protect participant confidentiality while maintaining the integrity necessary for secondary analyses exploring digital technology use and mental health relationships among emerging adults.

Physiological and emotional assessment of college students using wearable and mobile devices during the 2020 COVID-19 lockdown: An intensive, longitudinal dataset

This dataset was collected from university students before, during, and after the COVID-19 lockdown in Southern California. Data collection happened continuously for the average of 7.8 months (SD=3.8, MIN=1.0, MAX=13.4) from a population of 21 students of which 12 have also completed an exit survey, and 7 started before the California COVID-19 lockdown order. This multimodal dataset included different means of data collection such as Samsung Galaxy Watch, Oura Ring, a Life-logger app named Personicle, a questionnaire mobile app named Personicle Questions, and periodical and personalised surveys. The dataset contains raw data from Photoplethysmogram (PPG), Inertial measurement unit (IMU), and pressure sensors in addition to processed data on heart rate, heart rate variability, sleep (bedtime, sleep stages, quality), and...

ABSTRACT: With the popularization of low-cost mobile and wearable sensors, prior studies have utilized such sensors to track and analyze people's mental well-being, productivity, and behavioral patterns. However, there still is a lack of open datasets collected in-the-wild contexts with affective and cognitive state labels such as emotion, stress, and attention, which would limit the advances of research in affective computing and human-computer interaction. This work presents K-EmoPhone, an in-the-wild multi-modal dataset collected from 77 university students for seven days. This dataset contains (i) continuous probing of peripheral physiological signals and mobility data measured by commercial off-the-shelf devices; (ii) context and interaction data collected from individuals' smartphones; and (iii) 5,582 self-reported affect states, such as emotion, stress, attention, and disturbance, acquired by the experience sampling method. We anticipate that the presented dataset will contribute to the advancement of affective computing, emotion intelligence technologies, and attention management based on mobile and wearable sensor data.

Last update: Apr. 12, 2023

• Version 1.1.2 (Jun. 3, 2023)

Published the dataset at Scientific Data Journal.

Updated end-user license agreement.

• Version 1.1.1 (Apr. 12, 2023)

Updated file description and abstract.

• Version 1.1.0 (Feb. 5, 2023)

Updated the quality of the sensor data information.

Deleted three participants (P27, P59, P65) due to the low quality issue.

• Version 1.0.0 (Aug. 3, 2022)

Added P##.zip files, where each P## means the separate participant.

Added SubjData.zip file, which includes individual characteristics information and labels.

Quantitative and qualitative data sets for 24 sites across Ghana, Malawi and South Africa:
a) SPSS dataset on young people’s use of mobile phones in Ghana, Malawi and South Africa.  4626 cases (young people aged 7-25 years): 1568 Ghana; 1544 Malawi; 1514 South Africa.  719 variables (+ 11 ‘navigation facilitators’) b) 1,620 Qualitative transcripts from interviews with people of diverse ages, 8y upwards: individual interviews [using either i.theme checklist or ii call register checklist]; focus group interviews [not all sites]: 50-80 transcripts for most sites.

This research project, which commenced in August 2012, explored how the rapid expansion of mobile phone usage is impacting on young lives in sub-Saharan Africa. It builds directly on our previous research on children’s mobility within which baseline quantitative data and preliminary qualitative information was collected on mobile phone usage (2006-2010) across 24 research sites, as an adjunct to our wider study of children’s physical mobility and access to services.

In this study our focus is specifically on mobile phones and we cover a much wider range of phone-related issues, including changes in gendered and age patterns of phone use over time; phone use in building social networks (for instance to support job search); impacts on education, livelihoods, health status, safety and surveillance, physical mobility and possible connections to migration, youth identity, and questions of exploitation and empowerment associated with mobile phones.

Mixed-method, participatory youth-centred studies have been conducted in the same 24 sites as in our earlier work across Ghana, Malawi and South Africa (urban, peri-urban, rural, remote rural, in two agro-ecological zones per country). We have built on the baseline data for 9-18 year-olds gathered in 2006-2010, through repeat and extended studies, but also included additional studies with 19-25 year-olds (to capture changing usage and its impacts as our initial cohort move into their 20s).

BackgroundWith the ongoing development of the information society, the Internet and smartphones have become an essential way of life, but also fostered the problem of mobile phone dependence. Physical activity and subjective well-being have both been shown to correlate with mobile phone dependence, but the impact of subjective well-being on the relationship between physical activity and mobile phone dependence is not fully understood. This study investigates subjective well-being as a potential mediating variable in the relationship. It also investigates whether psychological capital moderates the association between subjective well-being and mobile phone dependence.MethodsA total of 9,569 students from 38 universities in Jiangsu Province were selected. Participants were surveyed via the online questionnaire distribution platform Questionnaire Star. Common method bias test and Pearson correlation tests were used to analyze the study indicators, and the theoretical model for this study was validated using Process plug-in developed by Hayes and set at p < 0.05 (two- tail) as statistically significant.ResultsThe levels of physical activity, subjective well-being, and psychological capital were all significantly higher for male students than female students. However, female students had a significantly higher level of mobile phone dependence. As predicted, there was a direct negative correlation between physical activity and mobile phone dependence, and subjective well-being mediated that relationship. Psychological capital moderated the relationship between subjective well-being and mobile phone dependence. It also positively moderated the indirect effect between physical activity and mobile phone dependence via subjective well-being.

BackgroundPsychological problems often occur in college students, with the most common ones being depression and anxiety symptoms. Exploring the risk factors that influence depression and anxiety symptoms in college students is essential to promote their physical and mental health.ObjectiveThis study aimed to investigate the independent and interaction effects of problematic mobile phone use (PMPU) and the number of close friends (NCFs) on depression and anxiety symptoms and the comorbidity of these symptoms among college students.MethodsA cross-sectional survey was conducted in Huainan, Anhui Province, and Suzhou, Jiangsu Province in China from October to December 2022. Data from 7,617 college students were collected. The Patient Health Questionnaire and Generalized Anxiety Disorder-7 were used to evaluate depression and anxiety symptoms. The PMPU data were collected by the Mobile Phone Addiction Type Scale. Multinomial logistic regression models were performed to examine the associations of PMPU and NCFs with depression and anxiety symptoms and their interaction effects.ResultsPMPU and lack of close friends significantly increased the risk of depression and anxiety symptoms and the comorbidity of these symptoms in college students (p 

Dataset Source

The dataset is from the University of Ottawa Link: http://nextconlab.academy/MCSData/MCS-FakeTaskDetection.html

Description

• Dataset is generated by CrowdSenSim simulation tool
• Dataset contains legitimate tasks and fake tasks
• The task attributed are as follows: {’ID’, ’latitude’, ’longitude’, ’day’, ’hour’, ’minute’, ’duration’, ’remaining time’, ’battery requirement %’, ’Coverage’, ’legitimacy’, ’GridNumber’, ’OnpeakHour’}.
• Location of tasks are specified by ’latitude’ and ’longitude’ together.
• ’day’, ’hour’ and ’minute’ describe the task publish time.
• ’Duration’ denotes task active duration in terms of minutes.
• ’Remaining time’ denotes the residual time of a sensing task till its completion.
• ’Battery requirement’ is percentage of battery required to complete a task.
• ’Coverage’ denotes task sensing distance.
• ’Legitimacy’ describes whether a task is illegitimate one or legitimate one. This feature is used only in training of the machine learning models as the MCS platform is unaware of task legitimacy when a task is submitted.
• ’GridNumber’ is obtained by splitting sensing city map to small grids with numbers beginning at 1.
• ’OnpeakHour’ is a binary flag to indicate if task start time occurs during 7am to 11am. We define 7am to 11am as the peak hour and other hours are non-peak for the sake of simplicity in simulations.
• The dataset is created including total 14,484 tasks, with 12,587 legitimate tasks and 1,897 fake tasks, respectively.

With the rapid development of Internet technology, more and more college students are facing the threat of mobile phone addiction. However, the relationship and underlying mechanism between mobile phone addiction and academic burnout haven’t been explored in depth. This study proves the mediating role of technology conflict and the moderating role of mindfulness in the relation between mobile phone addiction and academic burnout. 752 college students were recruited to complete the questionnaire of mobile phone addiction, technology conflict, mindfulness and academic burnout. Results showed that mobile phone addiction was significantly and positively associated with academic burnout, and this relationship could be mediated by technology conflict. Besides, the direct effect of mobile phone addiction on academic burnout and the indirect effect of technology conflict in this link were moderated by mindfulness. Both these two effects are stronger for college students with lower level of mindfulness. Our findings enrich our understanding of how and when mobile phone addiction was related to academic burnout. Educational professionals and parents should take timely measure to the academic burnout of college students suffering from mobile phone addiction, particularly for those with lower level of mindfulness.

LMOS_TraceGas_SurfaceMobile_UWEC-Auto_Data_1 is the Lake Michigan Ozone Study (LMOS) trace gas surface mobile data collected onboard the University of Wisconsin-Eau Claire (UWEC) surface mobile platform during the LMOS field campaign. This product is a result of a joint effort across multiple agencies, including NASA, NOAA, the EPA, Electric Power Research Institute (EPRI), National Science Foundation (NSF), Lake Michigan Air Directors Consortium (LADCO) and its member states, and several research groups at universities. Data collection for this product is complete.Elevated spring and summertime ozone levels remain a challenge along the coast of Lake Michigan, with a number of monitors recording levels/amounts exceeding the 2015 National Ambient Air Quality Standards (NAAQS) for ozone. The production of ozone over Lake Michigan, combined with onshore daytime “lake breeze” airflow is believed to increase ozone concentrations at locations within a few kilometers off shore. This observed lake-shore gradient motivated the Lake Michigan Ozone Study (LMOS). Conducted from May through June 2017, the goal of LMOS was to better understand ozone formation and transport around Lake Michigan; in particular, why ozone concentrations are generally highest along the lakeshore and drop off sharply inland and why ozone concentrations peak in rural areas far from major emission sources. LMOS was a collaborative, multi-agency field study that provided extensive observational air quality and meteorology datasets through a combination of airborne, ship, mobile laboratories, and fixed ground-based observational platforms. Chemical transport models (CTMs) and meteorological forecast tools assisted in planning for day-to-day measurement strategies. The long term goals of the LMOS field study were to improve modeled ozone forecasts for this region, better understand ozone formation and transport around Lake Michigan, provide a better understanding of the lakeshore gradient in ozone concentrations (which could influence how the Environmental Protection Agency (EPA) addresses future regional ozone issues), and provide improved knowledge of how emissions influence ozone formation in the region.

ObjectiveTo investigate the effect of physical exercise on sleep quality and the mediating effect of smartphone use behavior in college students.MethodsA cross-sectional study design was adopted. An online survey of 5,075 college students was conducted using the Physical Activity Rating Scale-3, the Pittsburgh Sleep Quality Index, and the Mobile Phone Addiction Tendency Scale.ResultsThe sleep quality of college students was poor, and the proportion of college students with good sleep quality was 23.567%. A significant correlation existed between sleep quality and physical exercise (r = −0.159, P < 0.001) and mobile phone addiction (r = 0.355, P < 0.001). Physical exercise can predict sleep quality in college students (β = −0.011, P < 0.001). Smartphone use plays a part in mediating the process by which physical exercise affects sleep quality.ConclusionChinese college students have poor sleep quality. Physical exercise and smartphone use behavior are important factors affecting the sleep quality of college students. Physical exercise can directly predict the sleep quality of college students and can predict the sleep quality of college students through the mediating effect of smartphone use behavior.

Context (This dataset does not contain the RENAME2 File to avoid errors)

The Caltech 256 is considered an improvement to its predecessor, the Caltech 101 dataset, with new features such as larger category sizes, new and larger clutter categories, and overall increased difficulty. This is a great dataset to train models for visual recognition: How can we recognize frogs, cell phones, sail boats and many other categories in cluttered pictures? How can we learn these categories in the first place? Can we endow machines with the same ability? Content

There are 30,607 images in this dataset spanning 257 object categories. Object categories are extremely diverse, ranging from grasshopper to tuning fork. The distribution of images per category are:

Min: 80
Med: 100
Mean: 119
Max: 827

Acknowledgements

Original data source and banner image: http://www.vision.caltech.edu/Image_Datasets/Caltech256/

When using this dataset, please remember to cite:

Griffin, G. Holub, AD. Perona, P. The Caltech 256. Caltech Technical Report. Inspiration

Can you build a model that IDs certain images?
What is the object? Is it a backpack, chopsticks, fried egg, or one of the other 253 object categories?

During the present time, COVID-19 situation is the topmost priority in our life. We are introducing a new dataset named Covid Face-Mask Monitoring Dataset which is based on Bangladesh perspective. We have a main concern to detect people who are using masks or not in the street. Furthermore, few people are not wearing masks properly which is harmful for other people and we have the intention to detect them also. Our proposed dataset contains 6,550 images and those images collected from the walking street, bus stop, street tea stall, foot-over bridge and so on. Among the full dataset, we selected 5,750 images for training purposes and 800 images for validation purposes. Our selected dimension is 1080 × 720 pixels for entire dataset. The percentage of validation data from the full dataset is almost 12.20%. We used a personal cell phone camera, DSLR for collecting frames and adding them into our final dataset. We have also planned to collect images from the mentioned place using an action camera or CCTV surveillance camera. But, from Bangladesh perspective it is not easy to collect clear and relevant data for research. To extend, CCTV surveillance cameras are mostly used in the university, shopping complex, hospital, school, college where using a mask is mandatory. But our goal of research is different. In addition, we want to mention that in our proposed dataset there are three classes which are 1. Mask, 2. No_mask, 3. Mask_not_in_position.

TOLNet_UAH_Data is the lidar data collected by the Rocket-city O3 Quality Evaluation in the Troposphere (RO3QET) lidar at the University of Alabama in Huntsville, Alabama as part of the Tropospheric Ozone Lidar Network (TOLNet). Data collection for this product is ongoing.In the troposphere, ozone is considered a pollutant and is important to understand due to its harmful effects on human health and vegetation. Tropospheric ozone is also significant for its impact on climate as a greenhouse gas. Operating since 2011, TOLNet is an interagency collaboration between NASA, NOAA, and the EPA designed to perform studies of air quality and atmospheric modeling as well as validation and interpretation of satellite observations. TOLNet is currently comprised of seven Differential Absorption Lidars (DIAL). Each of the lidars are unique, and some have had a long history of ozone observations prior to joining the network. Five lidars are mobile systems that can be deployed at remote locations to support field campaigns. This includes the Langley Mobile Ozone Lidar (LMOL) at NASA Langley Research Center (LaRC), the Tropospheric Ozone (TROPOZ) lidar at the Goddard Space Flight Center (GSFC), the Tunable Optical Profile for Aerosol and oZone (TOPAZ) lidar at the NOAA Chemical Sciences Laboratory (CSL) in Boulder, Colorado, the Autonomous Mobile Ozone LIDAR instrument for Tropospheric Experiments (AMOLITE) lidar at Environment and Climate Change Canada (ECCC) in Toronto, Canada, and the Rocket-city O3 Quality Evaluation in the Troposphere (RO3QET) lidar at the University of Alabama in Huntsville, Alabama. The remaining lidars, the Table Mountain Facility (TMF) tropospheric ozone lidar system located at the NASA Jet Propulsion Laboratory (JPL), and City College of New York (CCNY) New York Tropospheric Ozone Lidar System (NYTOLS) are fixed systems.TOLNet seeks to address three science objectives. The primary objective of the network is to provide high spatio-temporal measurements of ozone from near the surface to the top of the troposphere. Detailed observations of ozone structure allow science teams and the modeling community to better understand ozone in the lower-atmosphere and to assess the accuracy and vertical resolution with which geosynchronous instruments could retrieve the observed laminar ozone structures. Another objective of TOLNet is to identify an ozone lidar instrument design that would be suitable to address the needs of NASA, NOAA, and EPA air quality scientists who express a desire for these ozone profiles. The third objective of TOLNET is to perform basic scientific research into the processes create and destroy the ubiquitously observed ozone laminae and other ozone features in the troposphere. To help fulfill these objectives, lidars that are a part of TOLNet have been deployed to support nearly ten campaigns thus far. This includes campaigns such as the Deriving Information on Surface conditions from Column and Vertically Resolved Observations Relevant to Air Quality (DISCOVER-AQ) mission, the Korea United States Air Quality Study (KORUS-AQ), the Tracking Aerosol Convection ExpeRiment – Air Quality (TRACER-AQ) campaign, the Front Range Air Pollution and Photochemistry Éxperiment (FRAPPÉ), the Long Island Sound Tropospheric Ozone Study (LISTOS), and the Ozone Water–Land Environmental Transition Study (OWLETS).

Introduction

The Mobile phone activity dataset is composed by one week of Call Details Records (CDRs) from the city of Milan and the Province of Trentino (Italy).

Description of the dataset

Every time a user engages a telecommunication interaction, a Radio Base Station (RBS) is assigned by the operator and delivers the communication through the network. Then, a new CDR is created recording the time of the interaction and the RBS which handled it. The following activities are present in the dataset:

• received SMS
• sent SMS
• incoming calls
• outgoing calls
• Internet activity

In particular, Internet activity is generated each time a user starts an Internet connection or ends an Internet connection. Moreover, during the same connection a CDR is generated if the connection lasts for more than 15 min or the user transferred more than 5 MB.

The datasets is spatially aggregated in a square cells grid. The area of Milan is composed of a grid overlay of 1,000 (squares with size of about 235×235 meters. This grid is projected with the WGS84 (EPSG:4326) standard. For more details we link the original paper http://go.nature.com/2fcOX5E

The data provides CellID, CountryCode and all the aforementioned telecommunication activities aggregated every 60 minutes.

Original datasource

The Mobile phone activity dataset is a part of the Telecom Italia Big Data Challenge 2014, which is a rich and open multi-source aggregation of telecommunications, weather, news, social networks and electricity data from the city of Milan and the Province of Trentino (Italy). The original dataset has been created by Telecom Italia in association with EIT ICT Labs, SpazioDati, MIT Media Lab, Northeastern University, Polytechnic University of Milan, Fondazione Bruno Kessler, University of Trento and Trento RISE. In order to make it easy-to-use, here we provide a subset of telecommunications data that allows researchers to design algorithms able to exploit an enormous number of behavioral and social indicators. The complete version of the dataset is available at the following link: http://go.nature.com/2fz4AFr

Relevant, external, data sources

The presented datasets can be enriched by using census data provided by the Italian National Institute of Statistics (ISTAT) (http://www.istat.it/en/), a public research organization and the main provider of official statistics in Italy. The census data have been released for 1999, 2001 and 2011. The dataset (http://www.istat.it/it/archivio/104317), released in Italian, is composed of four parts: Territorial Bases (Basi Territoriali), Administrative Boundaries (Confini Amministrativi), Census Variables (Variabili Censuarie) and data about Toponymy (Dati Toponomastici).

Motivational video: https://www.youtube.com/watch?v=_d2_RWMsUKc

Relevant papers

Blondel, Vincent D., Adeline Decuyper, and Gautier Krings. "A survey of results on mobile phone datasets analysis." EPJ Data Science 4, no. 1 (2015): 1.

Francesco Calabrese, Laura Ferrari, and Vincent D. Blondel. 2014. Urban Sensing Using Mobile Phone Network Data: A Survey of Research. ACM Comput. Surv. 47, 2, Article 25 (November 2014), 20 pages.

Eagle, Nathan, Michael Macy, and Rob Claxton. "Network diversity and economic development." Science 328, no. 5981 (2010): 1029-1031.

Lenormand, Maxime, Miguel Picornell, Oliva G. Cantú-Ros, Thomas Louail, Ricardo Herranz, Marc Barthelemy, Enrique Frías-Martínez, Maxi San Miguel, and José J. Ramasco. "Comparing and modelling land use organization in cities." Royal Society open science 2, no. 12 (2015): 150449.

Louail, Thomas, Maxime Lenormand, Oliva G. Cantu Ros, Miguel Picornell, Ricardo Herranz, Enrique Frias-Martinez, José J. Ramasco, and Marc Barthelemy. "From mobile phone data to the spatial structure of cities." Scientific reports 4 (2014).

De Nadai, Marco, Jacopo Staiano, Roberto Larcher, Nicu Sebe, Daniele Quercia, and Bruno Lepri. "The Death and Life of Great Italian Cities: A Mobile Phone Data Perspective." WWW, 2016.

Citation

We kindly ask people who use this dataset to cite the following paper, where this aggregation comes from:

Barlacchi, Gianni, Marco De Nadai, Roberto Larcher, Antonio Casella, Cristiana Chitic, Giovanni Torrisi, Fabrizio Antonelli, Alessandro Vespignani, Alex Pentland, and Bruno Lepri. "A multi-source dataset of urban life in the city of Milan and the Province of Trentino." Scientific data 2 (2015).

Dataset description

The dataset description will start with describing the local conditions and other metadata, then will continue with describing the recording procedure and annotation methodology. Finally, a brief description of the dataset deployment and publication will be given.

Meta Information

The dataset was recorded at King's College London (KCL) Hospital, Denmark Hill, Brixton, London SE5 9RS in the period from 26 to 29 September 2017. We used a typical examination room with about ten square meters area and a typical reverberation tome of approx. 500ms to perform the voice recordings. Due to the fact, that the voice recordings are performed in the realistic situation of doing a phone call (i.e. participant holds the phone to the preferred ear and microphone is in direct proximity to the mouth), one can assume that all recordings were performed within the reverberation radius and thus can be considered as “clean”.

Recording Procedure

We used a Motorola Moto G4 Smartphone as recording device. To perform the voice recordings on the device, we developed a “Toggle Recording App”, which uses the same functionalities as the voice recording module used within the i-PROGNOSIS Smartphone application, but deployed as a standalone android application. This means, that the voice capturing service runs as a standalone background service on the recording device and triggers voice recordings via on- and off-hook signals of the Smartphone. Due to the fact, that we directly record the microphone signal, and not the GSM (“Global System for Mobile Communications”) compressed stream, we end up with high quality recordings with a sample rate of 44.1 kHz and a bit depth of 16 Bit (audio CD quality). The raw, uncompressed data is directly written to the external storage of the Smartphone (SD-card) using the well-known WAVE file format (.wav). We used the following workflow to perform a voice recording:

Ask the participant to relax a bit and then to make a phone call to the test executor (off-hook signal triggered).}

Ask the participant to read out “The North Wind and the Sun”

Depending on the constitution of the participant either ask to read out “Tech. Engin. Computer applications in geography snippet”

Start a spontaneous dialog with the participant, the test executor starts asking random questions about places of interest, local traffic, or personal interests if acceptable.

Test executor ends call by farewell (on-hook signal triggered).

Annotation Scheme

For each HC and PD participant, we labeled the data regarding scores on the Hoehn & Yahr (H&Y), as well as the UPDRS II part 5 and UPDRS III part 18 scale. The voice recordings are labeled in the following scheme:

SI_ HS_ HYR_ UPDRS II-5_UPDRS III-18

with

SI as subject identification in the form IDNN, N in [0, 9]

HS as the health status label (hc or pd accordingly)

HYR as the expert assessed H&Y scale rating

UPDRS II-5 as the according expert peer-reviewed score

UPDRS III-18 as the according expert assessed score

For example, an audio recording with the file name “ID02_pd_1_2_1.wav” represents a recording of the third participant (First participant was anonymized as ID00), which has PD and a H&Y rating of 1, a UPDRS II-5 score of 2 and a UPDRS III-18 score of 1. At this point, it should be noted, that also all healthy controls were evaluated with regard to the introduced scales, because Parkinson's disease and voice degradation correlate, but don't match exactly. This means, that the data set includes one HC participant (ID31) with UPDRS II-5 and III-18 rating of 1, and also includes PD patients with UPDRS II-5 and III-18 ratings of 0. It should be emphasized, that this does not mean the data set includes ambiguous information, but that an expert was not able to hear voice degradation that would end up in a UPDRS rating greater than zero. Machine learning approaches may be able to nevertheless classify correctly, or at least learn to correlate, but not match PD and voice degradation at any time.

Appendix

North Wind and the Sun (Orthographic Version):

“The North Wind and the Sun were disputing which was the stronger, when a traveler came along wrapped in a warm cloak. They agreed that the one who first succeeded in making the traveler take his cloak off should be considered stronger than the other. Then the North Wind blew as hard as he could, but the more he blew the more closely did the traveler fold his cloak around him; and at last the North Wind gave up the attempt. Then the Sun shone out warmly, and immediately the traveler took off his cloak. And so the North Wind was obliged to confess that the Sun was the stronger of the two.”

BNC – Tech. Engin. Computer applications in geography snippet:

“[...] This is because there is less scattering of blue light as the atmospheric path length and consequently the degree of scattering of the incoming radiation is reduced. For the same reason, the sun appears to be whiter and less orange-coloured as the observer's altitude increases; this is because a greater proportion of the sunlight comes directly to the observer's eye. Figure 5.7 is a schematic representation of the path of electromagnetic energy in the visible spectrum as it travels from the sun to the Earth and back again towards a sensor mounted on an orbiting satellite. The paths of waves representing energy prone to scattering (that is, the shorter wavelengths) as it travels from sun to Earth are shown. To the sensor it appears that all the energy has been reflected from point P on the ground whereas, in fact, it has not, because some has been scattered within the atmosphere and has never reached the ground at all. [...]”

General Description

This dataset was developed from real data on the usage of the corporate data network at the Universidade Federal do Rio Grande do Norte (UFRN). The main objective is to enable detailed observation of the university's network infrastructure and make this data available to the academic community. Data collection started on August 30, 2023, with the last query conducted on February 7, 2025, covering a total of approximately 19 months of continuous observations. During this period, about 1.5 months of data were lost due to failures in the data collection process or maintenance of the system responsible for capturing the data.

The data collections cover administrative, academic, and classroom sectors, spanning a total of 13 buildings within the university, providing a broad view of the network across different environments.

The dataset contains a total of 1,675,843 entries, each with 49 attributes.

Dataset Attributes, by Category

1. Connected Machines and ARP (8 attributes)

• Number of Access, Wi-Fi, Security, and VoIP Machines: Indicates the number of machines connected to each type of network, providing insight into the network size and device load.
• ARP Value for Access, Wi-Fi, Security, and VoIP: Refers to the number of entries in the Address Resolution Protocol (ARP) table associated with each type of network. ARP is used to map IP addresses to MAC addresses and can indicate potential connectivity issues.

2. Traffic Metrics (18 attributes)

• Packet and Byte: Indicates whether the information queried is accounted in packets or transmitted bytes, with positive (1) or negative (-1) values.
• Downlink and Uplink Bandwidth by Packets (Access, Wi-Fi, Security, VoIP): Refers to the number of packets received or sent by devices connected to each network type.
• Downlink and Uplink Bandwidth by Bytes (Access, Wi-Fi, Security, VoIP): Refers to the number of bytes received or sent by devices connected to each network type.

3. Collection Context (5 attributes)

• Sector: The sector from which the data was collected (academic, administrative, or classroom).
• Date: The date of the data collection.
• Time of Day: The time period of the collection (morning, afternoon, or evening).
• Day of the Week: The day of the week when the collection occurred.
• Hour: The hour of the collection.

4. Asset Identification (4 attributes)

• Asset IP: The IP address of the monitored device.
• Asset Model: The model of the network device.
• Asset Part Number: The part number of the device.
• Asset Firmware: The firmware version in use on the device.

5. Asset Performance (6 attributes)

• CPU Usage (% - 1 min and 5 min): The percentage of CPU usage on the device in the last minute and the last five minutes.
• Memory Used (%): The percentage of memory used by the device.
• Total and Used Memory (Kb): The total amount and the used amount of memory on the device, measured in Kb.
• Temperature (°C): The temperature of the device in degrees Celsius.

6. Port Packet Metrics (8 attributes)

• Packet In and Out Counter: The number of packets of data that have entered and exited all the device's ports.
• Broadcast Packet In and Out Counter: The number of broadcast packets that have entered and exited all the device's ports.
• Multicast Packet In and Out Counter: The number of multicast packets that have entered and exited all the device's ports.
• Packet Error In and Out Counter: The number of error packets that have entered and exited all the device's ports.

Size and Format

The dataset contains approximately 1,675,843 entries, with 49 attributes per entry. It is available in CSV format.

The psychological and physiological health of undergraduates was correlated with the sleep quality, which can be improved through increasing physical activity. However, the correlations between physical activity and sleep quality are subject to various factors. In this study, we investigated the effects of self-control and mobile phone addiction on the correlations between physical activity on undergraduates’ sleep quality at the psychological and behavioral levels. Data was collected through a survey with a convenient sample of 2,274 students in China. The study utilized scales of physical activity, sleep quality, self-control, and mobile phone addiction to quantitatively evaluate the impact of physical activity on the sleep quality of undergraduates. The correlations were analyzed using SPSS 26.0, including descriptive statistics, confidence tests, common method bias tests, correlation analysis, and hypothesis tests. Pearson correlation analysis shows that physical activity was significantly correlated with sleep quality (r = -0.541, p < 0.001), and that physical activity and sleep quality were significantly correlated with self-control and mobile phone addiction. Regression analysis shows that physical activity had a significant positive regression effect on self-control (standardized regression coefficient β = 0.234, p < 0.001), a significant negative regression effect on mobile phone addiction (β = –0.286, p < 0.001), and a significant negative regression effect on sleep quality (β = –0.351, p < 0.001). Further, a chain mediation model of physical activity → self-control → mobile phone addiction → sleep quality was proposed. The findings provide basic data for college students to promote physical activity and improve sleep quality.

><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>

Welcome to the FEIS (Fourteen-channel EEG with Imagined Speech) dataset.

<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><

The FEIS dataset comprises Emotiv EPOC+ [1] EEG recordings of:

* 21 participants listening to, imagining speaking, and then actually speaking
 16 English phonemes (see supplementary, below)

* 2 participants listening to, imagining speaking, and then actually speaking
 16 Chinese syllables (see supplementary, below)

For replicability and for the benefit of further research, this dataset
includes the complete experiment set-up, including participants' recorded
audio and 'flashcard' screens for audio-visual prompts, Lua script and .mxs
scenario for the OpenVibe [2] environment, as well as all Python scripts
for the preparation and processing of data as used in the supporting
studies (submitted in support of completion of the MSc Speech and Language
Processing with the University of Edinburgh):

* J. Clayton, "Towards phone classification from imagined speech using
 a lightweight EEG brain-computer interface," M.Sc. dissertation,
 University of Edinburgh, Edinburgh, UK, 2019.

* S. Wellington, "An investigation into the possibilities and limitations
 of decoding heard, imagined and spoken phonemes using a low-density,
 mobile EEG headset," M.Sc. dissertation, University of Edinburgh,
 Edinburgh, UK, 2019.

Each participant's data comprise 5 .csv files -- these are the 'raw'
(unprocessed) EEG recordings for the 'stimuli', 'articulators' (see
supplementary, below) 'thinking', 'speaking' and 'resting' phases per epoch
for each trial -- alongside a 'full' .csv file with the end-to-end
experiment recording (for the benefit of calculating deltas).

To guard against software deprecation or inaccessability, the full repository
of open-source software used in the above studies is also included.

We hope for the FEIS dataset to be of some utility for future researchers,
due to the sparsity of similar open-access databases. As such, this dataset
is made freely available for all academic and research purposes (non-profit).

><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>

REFERENCING

<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><

If you use the FEIS dataset, please reference:

* S. Wellington, J. Clayton, "Fourteen-channel EEG with Imagined Speech
 (FEIS) dataset," v1.0, University of Edinburgh, Edinburgh, UK, 2019.
 doi:10.5281/zenodo.3369178

><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>

LEGAL

<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><

The research supporting the distribution of this dataset has been approved by
the PPLS Research Ethics Committee, School of Philosophy, Psychology and
Language Sciences, University of Edinburgh (reference number: 435-1819/2).

This dataset is made available under the Open Data Commons Attribution License
(ODC-BY): http://opendatacommons.org/licenses/by/1.0

><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>

ACKNOWLEDGEMENTS

<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><

The FEIS database was compiled by:

Scott Wellington (MSc Speech and Language Processing, University of Edinburgh)
Jonathan Clayton (MSc Speech and Language Processing, University of Edinburgh)

Principal Investigators:

Oliver Watts (Senior Researcher, CSTR, University of Edinburgh)
Cassia Valentini-Botinhao (Senior Researcher, CSTR, University of Edinburgh)

<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><

METADATA

><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>

For participants, dataset refs 01 to 21:

01 - NNS
02 - NNS
03 - NNS, Left-handed
04 - E
05 - E, Voice heard as part of 'stimuli' portions of trials belongs to
   particpant 04, due to microphone becoming damaged and unusable prior to
   recording
06 - E
07 - E
08 - E, Ambidextrous
09 - NNS, Left-handed
10 - E
11 - NNS
12 - NNS, Only sessions one and two recorded (out of three total), as
   particpant had to leave the recording session early
13 - E
14 - NNS
15 - NNS
16 - NNS
17 - E
18 - NNS
19 - E
20 - E
21 - E

E = native speaker of English
NNS = non-native speaker of English (>= C1 level)

For participants, dataset refs chinese-1 and chinese-2:

chinese-1 - C
chinese-2 - C, Voice heard as part of 'stimuli' portions of trials belongs to
      participant chinese-1

C = native speaker of Chinese

<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><

SUPPLEMENTARY

><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>

Under the international 10-20 system, the Emotiv EPOC+ headset 14 channels:

F3 FC5 AF3 F7 T7 P7 O1 O2 P8 T8 F8 AF4 FC6 F4

The 16 English phonemes investigated in dataset refs 01 to 21:

/i/ /u:/ /æ/ /ɔ:/ /m/ /n/ /ŋ/ /f/ /s/ /ʃ/ /v/ /z/ /ʒ/ /p /t/ /k/

The 16 Chinese syllables investigated in dataset refs chinese-1 and chinese-2:

mā má mǎ mà mēng méng měng mèng duō duó duǒ duò tuī tuí tuǐ tuì

All references to 'articulators' (e.g. as part of filenames) refer to the
1-second 'fixation point' portion of trials. The name is a layover from
preliminary trials which were modelled on the KARA ONE database
(http://www.cs.toronto.edu/~complingweb/data/karaOne/karaOne.html) [3].

<>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <>< <><
><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><> ><>

[1] Emotiv EPOC+. https://emotiv.com/epoc. Accessed online 14/08/2019.

[2] Y. Renard, F. Lotte, G. Gibert, M. Congedo, E. Maby, V. Delannoy,
  O. Bertrand, A. Lécuyer. “OpenViBE: An Open-Source Software Platform
  to Design, Test and Use Brain-Computer Interfaces in Real and Virtual
  Environments”, Presence: teleoperators and virtual environments,
  vol. 19, no 1, 2010.

[3] S. Zhao, F. Rudzicz. "Classifying phonological categories in imagined
  and articulated speech." In Proceedings of ICASSP 2015, Brisbane
  Australia, 2015.