Smiley face
Smiley face

WeAreUs Mobile App

The WeAreUs app is a reflection tool empowering you to discover the people who most interrupt your life, for what reason and how these notifications add additional tasks to your life.

The app helps visualise the prominent stories that are occurring in your life at a given moment and what contacts of yours these stories involve. On reflection, you can identify whether the notifications occupying your time are adding value to the goals you have set out for your life. Time is precious, don't waste it dealing with unnecessary notifications.

The goal of this app is to help you identify interruptions which are worthy of your attention and those not so much.

This app hopes to grow into an intelligent manager of your notifications, alerting you only when necessary, giving you the opportunity to be present in the world and focus on things that really matter.

Research Perspective

The WeAreUs app includes a research study which we urge you to participate in. The intelligence empowering the app is only as good as the research behind it which means: the more participation in this study, the better the app gets! Simples.

App Functions

import_contacts

Stories

Creates abstract stories based on your mobile notifications. Stories contain the people, subject category and representative images of your notifications. They give you a snapshot of the important subjects in your life and with which contacts you share them with.

whatshot

Top 5

Identifies the top 5 contacts from whom you accept the most mobile notification interruptions. These are important people in your life. Jim Rohn said - We are the average of the 5 people we spend most time with. In this modern era, it holds true not just for our physical lives, but our digital lives too.

We also identify your top 5 contacts in different places. For example, you may have different Top 5's depending on where you are: at college, in the park, at a train station.

assignment

To-Do (Coming soon)

We can't always respond or take action on our notifications - WeAreUs identifies the task a notification might represent and adds it to a todo list which will notify you when it's relevant.

At the moment this feature populates a list of notifications which it determines will generate a task for you to do. For example, a WhatsApp message or Email that requires a reply, a social notification that requires an action etc.

person_pin_circle

I am Present (Coming soon!)

This feature enables the WeAreUs app to manage which notifications are permitted to disrupt you. The app monitors incoming notifications and determines whether or not you would open it. If not, it saves it and alerts you at an opportune time. This feature encourages you to be Present in the world - safe in the knowledge that if a notification really needs your attention, the app will alert you.

If you really wish to see if you have received a notification, they are listed within the app, hence you don't miss anything if you want to be disturbed. If the sender of a notification also has the WeAreUs app, they will be notified if you are currently in Present mode - so they won't be expecting a timely reply if it is not urgent.

schedule

They are Present (Coming soon!)

Our app can provide predictions regarding the time it will take your contacts to act upon a notification you send them at a given moment. This is to encourage you to put away your smartphone instead of waiting for a reply.

It can also predict the notification subject topics your contacts are generally open to accepting at a given moment. This is to disuade you from sending them notifications which do not align with their contextual desires. If it doesn't add value, don't send it!

school

Hat Tip

If you appear in a contacts top 5 contact list, you deserve a Hat Tip as they accept a lot of your notifications out of the masses of notifications - your notifications add value to their lives. We populate a list of your contacts who have you in their top 5, giving you a Hat Tip!

visibility

Sneak Peek

Some notifications hide the full message content. Sometimes this is in order to entice you to click and read further other times it's simply because there is so much text it runs out of space. This feature enables you to read the full content of a notification without alerting the app which posted it. For example, if you wish to read a WhatsApp message without alerting the sender you have read it.

record_voice_over

NAbs Amazon Alexa Skill (Coming soon!)

Paired with the NAbs Alexa skill, using the WeAreUs app enables you to ask Alexa to summarize your current active notifications, read in detail the active notifications content or remove active notifications - completely through spoken conversational commands and without touching your phone! All to encourage you to leave your phone behind, and be present!

Motivation

  • A short poem by Max Stossel nad Sander van Dijk which reflects on how we spend our time in the digital age of attention seeking technology.
  • Ed Sheeran favours an old Nokia "smart" phone which won't disturb him as he appreciates the world around him.
  • Inspiring video highlighting all the wonders and possibilities we miss when we have our heads stuck in our smartphones.
  • Simon Sinek talks about our real-world relationships and how our current behaviours with smartphones neglect their growth.
  • Michael McIntyre pokes fun at how smartphones have changed us.
  • Simon Sinek explains the benefits of limiting our smartphone usage in order to let our subconscious brain do what it does best - solve our problems and boost our creative prowess.
  • Thought provoking video challenging us to rethink our over-use of technology and social networks.

Research Study Participation

Submit your email on the Interest page and we will get you up and running in the study as soon as possible.

The following sections outline a general guide to particpation in this research study. It is advised that you read each section carefully before proceeding to participate in the study.

Background

Through participation in this study you will be contributing to research which aims to develop a framework which manages mobile notifications intelligently on behalf of a user. By engaging with the WeAreUs mobile application and answering short questionnaires when prompted, you will be enabling research which strives to better understand behavior toward information which is pushed at consumers without regard for their context and/or preference. This study also aims to uncover patterns of tasks which are generated from incoming mobile notifications to ascertain what notifications need to be dealt with directly by the consumer and what could potentially be dealt with autonomously on their behalf.

Data Collected

The purpose of this study is to understand the behavior surrounding mobile notifications and their intended consumer. Therefore, by participating in this study, you consent to the WeAreUs application logging details of your mobile notifications and monitoring some of your mobile usage statistics throughout the study. The details of the mobile notification are abstracted and are not stored until approved by you, the participant. For example, a notification sent by John asking about the results of a football match would be abstracted to: a ‘friend’ sent a ‘social’ notification. Other information is also recorded through questionnaires which ask you to reflect on certain aspects of received notifications. A description of all data points being logged can be found here. You are encouraged to check what is being collected through the application – all personal data collected is secured and pseudonymized to ensure continued privacy and prevention of identity discovery.

Getting Started and Useful tips

The procedure for engagement with this study is as follows:

  • Download and install the WeAreUs mobile application for Android from the Google Play Store.
  • On opening the application for the first time you will encounter a welcome screen which will require some permissions to be granted (e.g. mobile usage statistics). Each permission asked of you is accompanied by a screen explaining the necessity of the permission being granted and how to proceed. If for any reason you do not wish to grant permission and continue, you have the option to leave.
  • On completion of granting permissions, you are prompted to create an account – on this screen you have the option to enter the research study (if you are over 18 years). If you have opted-in to the study, you will be taken to a consent screen which you must read and agree to before continuing. Again, you have the option to leave and opt-out.
  • Once an account is created you are free to explore the application. It visualises key stories and people from your mobile notifications and highlights who, through monitoring your acceptance of mobile notification interruptions, are most important to you.
  • As a member of the research study you will also receive occasional mobile notifications from the WeAreUs application which ask you to complete a short questionnaire on your experience and interaction with your mobile notifications. These responses are recorded for research purposes.
  • A number of mobile usage statistics are also observed and logged by the application in the background.
The procedure of removal from the study are as follows:
  • Navigate to the Connections tab in the WeAreUs application.
  • Scroll down to the Opt-in radio button and click.
  • A dialogue box will appear informing you that your data will no longer be collected. On this dialogue box you are also given the option to remove any previous data collected. Click okay for changes to take effect.
  • You will now have been removed from the study. You have the option to Opt-in again at any time by simply returning to the Connections tab and clicking the radio button once again.
Procedure of raising a complaint/providing feedback:
  • Navigate to the Connections tab of the WeAreUs application.
  • Scroll down to Provide Feedback.
  • You will be prompted to enter your feedback.
  • You can then submit, with the option of remaining anonymous.
Alternatively:
  • Navigate to http://www.weareus.eu/feedback in a browser.
  • Fill in the form provided.
  • You can then submit, with the option of remaining anonymous.

A leader of the research study will then be in touch if necessary.

Conflicts of Interest

The WeAreUs mobile application is released on the Google Play Store for download. Only participants who are 18 and above have the option of opting-in to the research study. Those who don’t can still use the application to visualize their mobile notifications however, no data is stored on them for research purposes. It is possible that a conflict of interest could occur during the study, but unlikely. This is because participants of the research study are not known to the researchers of the study. Each participant enters the research study at their own discretion through the mobile application, hence even if it’s known an individual has downloaded the application, it won’t be known whether they have entered the study unless they explicitly reveal it (which you are advised not to do). Once opted-in to the study, all personally identifiable information stored on participants for research purposes is pseudonymized which prevents researchers involved in the study (or anyone else) from identifying them. On completion of the study, further methods are used to ensure the data cannot be traced back to the participant.

Opt-out

Participation in this research study is completely voluntary and you can opt-out at any time, for any reason and without penalty. You are also free to delete recorded responses if you so wish through the WeAreUs mobile application (the process is described above).

Duration

It is recommended that you engage with the research study through the WeAreUs notification questionnaires for a minimum of two weeks. You shall be sent a maximum of 8 notifications per day, each of which should take a maximum of 1 minute to complete.

Risks

As with any entity that manages data, there are risks of malicious attacks and data being hacked. There are security measures in place to prevent this from happening, such as encryption and obfuscation. All information stored on the participant adheres to the rules outlined by the Data Protection Commissioner: https://www.dataprotection.ie/docs/Data-Protection-Rules/y/21.htm The app utilizes Google’s Natural Language Processing API to provide insights into the subject of your notifications, hence the text content of the notifications is sent to Google. Note that: I. Google will not use any of your content (such as text sent for analysis) for any purpose except to provide you with the Cloud Natural Language API service. II. Google won’t make the text sent available to the public or share it except to provide the service. III. Google stores the text for a short period of time in order to provide the service. Usually deleted within a few hours. Some request metadata is stored, such as size of request and time. IV. Google does not use the text sent to improve or train the service. V. Google does not claim ownership. VI. Google does ‘end-to-end’ protection and secures the data sent. The app also utilizes Google’s Map API to provide contextual insights into the places you receive notifications and how they affect delivery, therefore Google’s Privacy Policy governs how this data is used by Google. Google Privacy Policy As data is collected via a mobile application, it is possible that your account could be comprised if your phone is stolen. If you suspect your phone has been stolen you should contact a member of the research study so your WeAreUs account can be secured. Email: discoveru.weareus@gmail.com

Benefits

By participating in this research study, you will receive a deeper understanding of your behavior toward your mobile notifications. The WeAreUs application is a reflection tool empowering you to discover the people who most interrupt your life, for what reason and what extra tasks these notifications add to your life. In due course, additional features will be added which can aid in the management of your notifications alleviating your workload and removing unnecessary distractions.

Anonymity and Publication

The data collected in this study will be analysed and used for improving frameworks which aid toward managing mobile notifications. Therefore, findings may be published. In this case, your data will be treated with full confidentiality and will not be identified as yours. Personally Identifiable Information, such as participant id, are removed from the collected data set once you choose to opt-out of the study (the process of which is outlined above). The data shall not be traceable to individual participants. All further analysis, publication and presentation of findings will therefore not reference any individual participants by any identifying feature. For example, one such targeted conference for publication of result of findings is UbiComp 2018.

Cautions

We must advise that, while extremely unlikely, we are obligated to report any inadvertent discovery of illicit activities.

Research Data

The following sections outline the data points logged by the WeAreUs app and their necessity for research purposes. The data logged can be split into two main categories: 1) Information which is observed as a user interacts with their mobile on a daily basis; Note: this information is recorded in background services on the user's device. 2) Information which is obtained directly from the user through questionnaires.

As some of the data being logged is done so unobtrusively in the background, it is important to take a minute and read through the following list to ensure you are aware of what data is being collected.

User Data

Email: Your preferred email address. We won't contact you unless it is important.

Gender: This is used to analyse differences in notification management between genders.

Phone number: This is used to identify you and your device. It is also used to identify if you are in a contact list of another user of the app. If you are, and you are in their top 5 contacts, you receive a 'Hat Tip' for being a valuable contact to them.

Age: This is used to analyse differences in notification management between ages and also to ensure only 18+ users opt-in for research.

Name: This is used to simply personalise your app experience.

Opt-in: This is to identify whether or not you have opted-in to our research study. (Please do, it's beneficial to both us and you, promise!)

Intro: This identifies that you have read the intro slides to the app and understand how the app adds value to your life.

Consent: This is used to identify whether or not you have accepted our terms of the research study and whether you have given consent for us to include you in the study.

Facebook Contacts: You have the option to sync your Facebook contacts. By doing this, the app can match incoming notifications to more of your contacts and better understand and manage your notifications.

Twitter Contacts: You have the option to sync your Twitter contacts. By doing this, the app can match incoming notifications to more of your contacts and better understand and manage your notifications.

Phone Contacts: You have the option to sync your Phone contacts. By doing this, the app can match incoming notifications to more of your contacts and better understand and manage your notifications.

App Blacklist: This is used to identify apps which you don't want the app to analyse or log information on. By adding an app to this list, our app will completely ignore all notifications sent through this app.

Locations: This is used to identify the context in which you receive and act on notifications. The latitude and longitude of your location is not stored outside of your device. What is stored is the location category as defined by Google's Places API. For example: 'school' or 'park'.

Activities: This is used to identify the context in which your receive and act on notifications. The activity is defined by Google's activity recognition API. For example: 'walking', 'jogging'.

Contact Data

Phone: Your contact's phone number (if synced from your mobile contact list). This is only used to identify if your contact is also using the app. If so, we can check if you are in their top 5 list!

Name: Your contact's name is used to match them to incoming notifications.It is not shared.

Screen name: Some apps use screen names to identify contacts in notifications e.g. Twitter. We check if this could be used to match notifications to contacts also.

Associated App: This is the app the contact is associated with. For example, whether the contact came from someone you follow on Twitter or from your friend list on Facebook.

Photo: This is the url to your contact's profile photo.

Total Stories: Every time a notification is matched to a contact, a story is created for that contact. The total number of stories enables the app to highlight contacts who send you lots of notifications.

Total Accepted Stories: Like the total stories data point, this describes the total number of notifications that were successfully matched to a contact, but in contrast, only counts the stories spawned from notifications which you opened. This enables the app to highlight contacts whose incoming notifications added value to you.

App Id: If your contact is using the app also, their app id is added here. This enables the app to highlight to you which of your contacts use the app. You can then be extra careful when sending them notifications so as to break into their top 5!

Notification Data

Notify Id: This is the id provided by the posting app which specifies certain settings to be launched in the target app.

Key: This is the unique key for the notification instance.

App: This is the app which posted the notification.

Post time: This is the time the notification arrived to the notification bar on your phone.

Tag: This is the tag attached to the notification as supplied by the posting app.

Is Clearable: Whether or not the notification can be removed from the notification bar by the user.

Is Ongoing: Whether or not the notification is ongoing. For example, an app update notification.

Group Id: The id used to cluster notifications together in the notification bar. For example, Whatsapp groups notifications from different contacts together into one Whatsapp notification.

Sort Key: A key to order notifications amoung others from the same app.

Big Text: The longer text shown in big form in the notification

Compact Actions: The number of actions shown in the notification. For example, some notifications give you actions such as 'reply' or 'remind me later'.

Info Text: This is the small piece of additional text in a notification.

Picture: Whether or not a picture is included in the notification. Note: the actual picture is not saved!

Show Chronometer: Whether or not a count-up timer is shown in the notification.

Show When: WHether the 'when' time should be shown in the notification. The 'when' time is similar to the post time in that it is the time the notification arrived to the notification bar of the device.

Sub Text: The third line of text in the notification.

Summary Text: This is the summary info shown alongside the expanded notification.

Text: This is the main text payload of the notification.

Text Lines: This is a set of lines of text included in the notification.

Title: This is the title of the notification.

Title Big: text

Category: The category of the notification as defined by Android.

Colour: The ARGB colour of the notification background.

Icon Level: This is the number of icon levels in the notification.

LED ARGB: This is the colour the notification alert makes on your phone. For example, presently Whatsapp is white, SnapChat is yellow, Facebook Messenger is green.

Priority: This is the priority level of the notification as set by the posting app.

Vibrate: This is the vibration pattern used on arrival of the notification.

Visibility: This is identifies how and when the System UI of the device reveals the notification.

When Timestamp: This is similar to the Post Time. It indicates the time at which the notification arrived to the notification bar of the device.

Ticker Text: This is the text of the notification in unexpanded form.

Removed: This is the time at which the notification is removed from the notification bar (by either the user or by the device/app itself).

Clicked: Whether or not the notification was opened or dismissed.

Updates: The total number of times the notification was updated since arrival. For example, Whatsapp notifications are updated if a new message has been posted in the app and the notification hasn't been removed from the notification bar since the last message.

App Last Used: The time at which the app the notification was received through was last used/opened.

Seen Time: The time the notification was seen (this is assumed immediately if the notification arrived when the screen was on, otherwise it is the time at which the screen is next turned on).

Decision Time: The time between seeing the notification and acting upon it (opening or dismissing).

Response Time: The total time from the notification arriving on the device, to its removal.

Latitude and Longitude: The lat and lon of the phone on arrival of the notification. This value is not saved off-device. It is used simply to identify if the notification arrived at a user-defined location.

Keywords: These are entities extracted from the raw text of the notification (with people's names removed!)

Subject: This is the subject of the notification as defined by Google's Natural Language Processing API.

Location: Using the lat and lon of the notification, the app attempts to match the notification to a location in the user's managed list of locations (stored on device only!). The location name is then stored as 'home', or 'work' for example.

Associated Contacts: The number of associated contacts that were matched to the notification.

Activity: This value is derived using Google's Activity Recognition API, or failing that, from a user managaged list of activities from within the app.

Significant: This is an inferred field based on the number of times the app which posted the notification was launched and used. If the app is used more than average for the user, then the app is significant. If an app is rarely used, then it is deemed insignificant.

Place Category: This value is defined using the lat and lon of the device on arrival of the notification and Google's Places API. The category of the inferred place, such as 'school' or 'park' is saved to add context to the notification.

Mobile Usage Data

Mobile usage data is logged periodically in the background. The values and totals are stored on the device during a period (1-2 hours), are updated when an event (e.g. incoming notification) is detected and reset once a period has ended.

Notification Count: This is the total number of notifications received.

Notification Accepted Count: This is the total number of accepted (opened) notifications.

Average Notification Seen Time: This is the average time over a period that it takes the user to see that a notification has been delivered to the device. (Measured by logging when the device screen is turned on).

Average Notification Decision Time: This is the average time over a period that the user takes to act on a notification.

Average Notification Response Time: This is the average time over a period from a notification being delivered to the device to it being removed (opened or dismissed).

Unlock Count: This is the total number of times a user unlocks their phone over the period of time.

Click Count: This is the total number of times the user interacted with (touched) the device over the period of time.

Phone Usage Time: This is the total amount of time the user spent interacting with their phone over a period of time.

Total Time In Apps: This is the total amount of time a user spent in apps during a period of time.

Total App Launches: This is the total number of times apps were launched during a period of time.

Unique App Launches: This is the total number of unique apps that were launched in a period of time.

Total Significant Apps Launched: This is the total number of times significant apps were launched in a period of time. 'Significant' apps are apps which you use regularly.

Unique Significant Apps Launched: This is the total number of unique significant apps launched during a period of time.

Total Insignificant Apps Launched: This is the total number of times insignificant apps were launched in a period of time. 'Insignificant' apps are apps which you use rarely.

Unique Insignificant Apps Launched: This is the total number of unique insignificant apps launched during a period of time.

Call count: This is the number of calls received during a period of time.

Call Unique Count: This is the number of calls from unique contacts during a period of time.

Call Duration Average: This is the average duration time of all calls during a period of time.

SMS Count: This is the total number of SMS messages received during a period of time.

SMS Unique Count: This is the total number of SMS messages received from unique contacts during a period of time.

SMS Average Length: This is the average length of all SMS messages received during a period of time.

SMS Sent to Received Ratio: This is the ratio of sent SMS messages to received SMS messages during a period of time.

Time Since Last Sample: This is the time since the last sample of mobile usage data was taken.

Questionnaire Data

Participants of the research study receive notifications prompting them to complete a questionnaire if the following conditions are met:

  • They have recently removed a notification from the notification bar of their device - either through opening it, or dismissing it.
  • They have not answered a questionnaire in the past 2 hours.
  • They have not reached their quota of 8 answered questionnaires for that day.

Time Received: This is the time the notification which spawned the questionnaire arrived to the notification bar of the devce.

App: This is the app through which the notification arrived.

Action: This is the action the user took toward the notification (either opening it or dismissing it).

Sender Relationship: This is the relationship between the user has with the sender of the notification. For example, 'family', 'friend', 'app'.

Subject: This is the perceived subject of the notification content. This field is filled automatically using Google's Natural Language API - but the user can modify if it is incorrect.

Keywords: These are entities extracted from the message content of the notifications (people's names are removed!).

Location: This is the location of the device on notification arrival. An abstract location value is all that is stored. For example, 'home' or 'work'. Explicit coordinates are not stored.

Activity: This is the activity of the user on arrival of the notification. This is auto-populated using Google's Activity Recognition API, but can be modified if found to be incorrect.

Task Type: If the notification spawns a task for the user to complete, this is the type of task. For example, 'reply' to a message or 'meet' with a friend or 'search' for a new song.

Task App: If the notification spawns a task for the user to complete, this is the app(s) which they deem suitable in aiding them complete the task.

Task Deadline: If the notification spawns a task for the user to complete, this is the deadline the user wishes to have the task completed by.

Good Behaviour: This is a value out of 5 selected by the user as to whether their action toward this notification (opening or dismissing) can be deemed good behaviour. For example, opening a 'social' notification during a meeting might be deemed 'bad' behaviour, so a value of 1 given.

Behaviour Reason: This is the reason the user gives for selecting the value of 'Good Behaviour'. There are a preset list to choose from, or a user can add their own reason. For example, 'I was bored', 'got distracted by the alert', 'it aids a goal I have'.

Relevance: This is a value out of 5 selected by the user as to whether the notification which arrived was relevant at the given moment.

Relevance Evidence: This is the part of the notification which led the user to select the 'Relevance' value. For example, the 'sender' of the notification was the reason it was relevant, or the 'subject'.

Correct Device: This is a value signifying whether or not the notification was delivered to the best device for viewing by the user in their current context. For example, would it have been better delivered to the user's desktop computer, or their smartwatch.

Alternative Device: This is a list of devices which would have been better to deliver the notification to the user through.

Device Evidence: This is the part of the notification or user context which led the user to choose the 'Alternative Device'. For example, the 'activity' of the user was on a computer so the computer may have been a better choice.

Debrief

The following sections outline the use which your data will be put to on your completion of the research study.
Notification Delivery

A number of algorithms for intelligent mobile notification delivery are being developed which require real world data for growth and testing purposes. For example, implementation of a LSTM recurrent neural network, learning classifier system and association rule mining. By applying these algorithms it is hoped to derive a set of rules through which notifications can be intelligently delivered.

To-do List Generation

By understanding the type of tasks spawned by notifications, lists of to-do tasks can be managed for the user and perhaps automatically executed on their behalf.

Synthetic Dataset Generation

To lower the barrier for entry on development and testing notification management systems a shared synthetic mobile notification dataset is being developed which will enable all researchers to quickly benchmark test the performance of their system. The synthetic dataset requires seed data in order to grow and accurately imitate real world scenarios. The data collected in this study will be used to seed a synthetic dataset. Personally Identifiable Information (PII) will not be used in the synthetic dataset - instead the frequency of abstract terms will be used to generated a synthetic immitation of real world scenarios. If you completed a Myer-Briggs personality test, the results will be used to spawn datasets which immitate certain personalities.

Future Work

Future work will include incorporating user control into the notification management system so participants can actively engage with the rules generated on their behalf.

Contact

Name:
Kieran Fraser
Email:
discoveru.weareus@gmail.com; kieran.fraser@adaptcentre.ie

Register Interest

Sign up below and we will email you with more details on how you can get involved in our research study (or just play with the app).
Enter a valid email!