Abstract: Transcribed text and physiological data of a remote video conference participant are transmitted to a local device separately from the video data, which depicts the remote party during a time interval. An image of the video data is captured at a time instant within the time interval. A value of a remote party feature is determined remotely using the video data. The remote party feature can be the remote party's heart rate at the time instant. The value of the feature is received onto the local device. Audio data captures sounds spoken by the remote party and is converted by the remote device into words of text. The audio data converted into a particular word was captured at the time instant. The particular word is received onto the local device. The particular word and the value of the feature are displayed in association with one another on the local device.

Abstract: A method for determining a user's stress level is performed by a smartphone app. Touch and motion feature values are generated, the feature values are weighted by regression parameters, and a stress score is generated based on the weighted touch and motion feature values. The touch feature values indicate how the user's finger moves over the smartphone and are generated from touch data points including X positions, Y positions and associated touch timestamp values. The motion feature values indicate movement of the smartphone and are generated from motion data points including X movements, Y movements, Z movements and associated motion timestamp values. The regression parameters are generated using touch and motion data identified by other users as being acquired while those other users were experiencing various perceived levels of stress. The app indicates to the user whether the stress score is higher or lower than a previously generated stress score.

Abstract: An autocomplete function for textual input uses situational parameters to predict the next words the user is intending to type. Situational and temporal parameters are based on textual input and sensor data of the user. A past time window is based on the situational and temporal parameters. Historical textual input and sensor data during the time window relating to the situational parameters are retrieved from a storage device and aggregated. A pre-existing model that relates the situational parameter to the time window is used to select a situational value based on the textual input and sensor data. Words relating to the situational parameter are listed that the user is likely to input next based on the selected situational value. The words are ranked by the probability that the user is intending to type each of the words. The highest ranked word is displayed to the user on a user interface.

Abstract: A method for optimizing screen time for the user of a smartphone involves determines a target level of usage based on situational context. A period of time during which the user interacts with the smartphone is detected. A first operational mode is detected in which the user interacts with the smartphone during the period of time. A first situational context is identified in which the user is interacting with the smartphone. An interaction benefit to the user is determined based on the period of time, the first operational mode and the situational context. Whether the interaction benefit equals or exceeds the target level is determined. If the interaction benefit does not equal or exceed the target level, it is recommended that the user interact with the smartphone in a second operational mode or restrict the amount of time during which the user engages in the first operational mode.

Abstract: An autocomplete function for textual input uses situational parameters to predict the next words the user is intending to type. Situational and temporal parameters are based on textual input and sensor data of the user. A past time window is based on the situational and temporal parameters. Historical textual input and sensor data during the time window relating to the situational parameters are retrieved from a storage device and aggregated. A pre-existing model that relates the situational parameter to the time window is used to select a situational value based on the textual input and sensor data. Words relating to the situational parameter are listed that the user is likely to input next based on the selected situational value. The words are ranked by the probability that the user is intending to type each of the words. The highest ranked word is displayed to the user on a user interface.

Abstract: Video data captured during a time interval at the location of a remote party to a videoconference is received onto a remote device. The video data depicts the remote party. Audio data capturing sounds spoken by the remote party during the time interval is also received onto the remote device, which converts the audio data into words of text and captures prosodic information describing the sounds spoken by the remote party during the time interval. The words of text are received onto a local device. The prosodic information corresponding to the sounds spoken by the remote party during the time interval that were converted into the words of text are also received onto the local device. The words of text and prosodic information are stored in association with one another. A physiological parameter of the remote party is determined using the video data and is received onto the local device.

Abstract: Transcribed text and physiological data of a remote video conference participant are transmitted to a local device separately from the video data, which depicts the remote party during a time interval. An image of the video data is captured at a time instant within the time interval. A value of a remote party feature is determined remotely using the video data. The remote party feature can be the remote party's heart rate at the time instant. The value of the feature is received onto the local device. Audio data captures sounds spoken by the remote party and is converted by the remote device into words of text. The audio data converted into a particular word was captured at the time instant. The particular word is received onto the local device. The particular word and the value of the feature are displayed in association with one another on the local device.

Abstract: A method for recommending those interventions most likely to achieve a desired state involves predicting the efficacy and engagement of interventions based on the experience of prior users who undertook the interventions. Physiological and personal parameters of the user are acquired. The user's initial state and desired state are determined. The engagement and efficacy levels of each intervention are predicted and used to determine the likelihood that the transition achieved by each intervention achieves its predicted end state. The likelihood that a second transition achieves the desired state is also determined based on efficacy and engagement for the second transition whose starting state is the end state of the first transition. The first and second interventions are identified whose associated transitions have the greatest combined likelihood of achieving the desired state compared to all other intervention combinations. The user is then prompted to engage in the first and second interventions.

Abstract: A system for triggering mental healthcare services based on prediction of critical events receives both structured data and unstructured data about mental healthcare patients. Feature extraction is performed, thereby generating records of structured data, and strings of vectors of unstructured data. The system makes a quality assessment about the structured data, and a quality assessment about the unstructured data. In a model selection step, the system selects one model out of a plurality of selectable models, where the selection is made based on the quality assessments made. The selected model is trained with records of structured data and with strings of vectors of unstructured data. New real-time data is supplied to the trained model so that the trained model predicts whether a crisis event is likely to occur. If the system predicts that a crisis event is likely to occur, then the system outputs an alert.

Abstract: A system for determining a mental health state of a user and for adjusting output content accordingly includes a monitoring unit that monitors parameters of the user and acquires corresponding data, an analysis unit that extracts features from data acquired by the monitoring unit, a classification unit that detects change in the mental health state of the user based on the features extracted by the analysis unit and classifies the change in mental state of the user, a control unit that adapts the content to be output to the user by the output unit based on the detected change in the mental health state of the user classified by the classification unit, and an output unit that outputs content to the user based on the detected change in the mental health state of the user. A method determines the mental health state of the user and adjusts output content accordingly.

Abstract: An autocomplete function for textual input uses situational parameters to predict the next words the user is intending to type. Situational and temporal parameters are based on textual input and sensor data of the user. A past time window is based on the situational and temporal parameters. Historical textual input and sensor data during the time window relating to the situational parameters are retrieved from a storage device and aggregated. A pre-existing model that relates the situational parameter to the time window is used to select a situational value based on the textual input and sensor data. Words relating to the situational parameter are listed that the user is likely to input next based on the selected situational value. The words are ranked by the probability that the user is intending to type each of the words. The highest ranked word is displayed to the user on a user interface.

Abstract: A method for determining a user's stress level is performed by a smartphone app. Touch and motion feature values are generated, the feature values are weighted by regression parameters, and a stress score is generated based on the weighted touch and motion feature values. The touch feature values indicate how the user's finger moves over the smartphone and are generated from touch data points including X positions, Y positions and associated touch timestamp values. The motion feature values indicate movement of the smartphone and are generated from motion data points including X movements, Y movements, Z movements and associated motion timestamp values. The regression parameters are generated using touch and motion data identified by other users as being acquired while those other users were experiencing various perceived levels of stress. The app indicates to the user whether the stress score is higher or lower than a previously generated stress score.

Abstract: A method for detecting a user's outlier days uses data corresponding to features of the user acquired over multiple days by sensors on the user's electronic device. The data acquired for each day and feature is labeled as regular or irregular by applying N labeling approaches. One of the N labeling approaches compares the data for each feature with how values of previously acquired data for corresponding features are distributed. N labels are generated for the data for each feature and day. The machine learning classification model is trained using one of the N labels for each of the N labeling approaches. An optimal labeling approach is selected from among the N labeling approaches for each feature using the machine learning classification model. For each feature, the method determines whether each of the days is an outlier day for the user using the labels obtained with the optimal labeling approach.

Abstract: A user is validated using first data from sensors on the user's portable device and second data input by the user. The user's digital behavioral fingerprint is generated using the first data. Whether the user's purported identity is authentic is determined by computing an authenticity score using the fingerprint. The user's identity is authentic if the authenticity score is less than a first threshold and potentially not authentic if greater than the first threshold. If potentially not authentic, whether the second data is invalid is determined using a conformity score that compares the second data to prior data of the user and data associated with others. The second data is not invalid if the conformity score is less than a second threshold, but invalid if greater than the second threshold. Access to a service is granted if either the user's identity is authentic or the second data is not invalid.

Abstract: A method for detecting abnormal behavior involves constructing words and text documents based on data acquired from mobile phone sensors during defined time intervals. The time intervals are defined based on data from mobile phone sensors indicative of usage patterns of the mobile phone user. Words are constructed for each time interval as a vector including the time interval and sensor-based feature levels. Each sensor-based feature level is mapped to a range of values of a sensor-based feature that are extracted from the sensor data. The text document is constructed from the words based on the time intervals and the sensor-based feature levels. A current routine for each time interval is determined using topic modeling based on the words that most frequently appear in the text document. An alert is generated if the current routine for any time interval deviates from a past routine for a corresponding past time interval.

Abstract: A preventative healthcare system calibrates a risk model by assigning weights to attributes for the freshness, completeness and uncertainty of a user's medical information. A risk predictive model is implemented based on the medical information. The risk of a specific health outcome of the user is determined using the risk predictive model, which is calibrated by computing attribute scores for freshness, completeness and uncertainty of the medical information and by assigning weights to the attribute scores. A need-for-data (ND) score is computed using the weighted attribute scores. A need-for-checkup (NC) score is computed using traits of the user. The method determines that new medical information related to the user is needed or that the user needs a checkup based on the ND and NC scores. A prompt is delivered to the user indicating that new medical information related to the user is needed or that the user needs a checkup.

Abstract: A system for improving a user's wellbeing uses machine learning to evaluate statistical parameters and to recommend parameter ranges that will improve the user's wellbeing. The system receives user data including ambient and environmental data related to the user's location, physiological data related to the user's body, and behavioral data of the user. Statistical parameters are generated that characterize the data based on a time period corresponding to the data. A wellbeing score for the user is generated by applying a probabilistic model to the statistical parameters. The model determines a recommended range of values for the parameters related to the ambient data, environmental data, physiological data and behavioral data that will improve the user's wellbeing score based on the statistical parameters and the wellbeing score. Behavior modifications based on the recommended range of values for parameters are recommended to the user so as to improve the user's wellbeing score.

Abstract: A method for optimizing a user's behavioral changes uses an artificial intelligence system to present actions to the user for achieving a behavioral change goal. The goal is first set at a time instant t. A user characteristics vector and current state vector are generated using static and dynamic user characteristics describing the user's current state. The vectors are used to generate a behavioral change tool vector for the time instant t. The behavioral change tool vector is presented to the user to select a suggested action for achieving the goal. The user characteristics, current state, and behavioral change tool vectors for the time instant t are mapped to a next current state vector for a next time instant t+1. The system repeats generating the behavioral change tool vector using vectors updated for the next time instant t+1 until the mapping indicates that the behavioral change goal has been achieved.

Abstract: Method, system and computer program for determining personalized parameters for a user. The method comprises providing a first vector of personal characteristics based on received first data, a second vector of behavior and activity characteristics based on received second data, and a third vector of wellbeing measures based on received third data. Exhibited personal characteristics and the first vector is also calculated. A reference group for the user is created and a similarity measure between the user and the reference group is implemented to identify which of said users has more characteristics in common with the user. An optimal behavior and activity distribution vector can be determined from the most similar users of said reference group. The range of behaviors and activities that are good or bad for the user can be also determined.

Abstract: An electronic device includes a context scanner, an emotion feedback module, and a processing unit. The context scanner scans an area during a predetermined time period to detect position and proximity information of the electronic device of a user. The position information indicates the geographic location of the electronic device within the area. The proximity information indicates that the electronic device is in the proximity of a second electronic device of a second user. The emotion feedback module detects the user's emotional response during the time period by sensing physiological and physical signals of the user. The processing unit executes an algorithm that determines the relationship between the position information, the proximity information and the emotional response. The processing unit generates a report based on the relationship. The report indicates how the user's presence in the geographic location and the user's proximity to the second user influence the user's well-being.