Abstract: An apparatus includes a processor to: assign a portion of currently available instruction-based processing resources to a first non-neuromorphic performance of an analytical function; in response to availability of sufficient remaining processing resources for a first neuromorphic performance of the analytical function with the same input values, assign a portion of the remaining processing resources to the first neuromorphic performance; analyze the output values generated by the first neuromorphic and non-neuromorphic performances to determine a degree of accuracy of the neural network in performing the analytical function; in response to at least the degree of accuracy exceeding a predetermined threshold, assign a portion of currently available processing resources to a second neuromorphic performance of the analytical function; and in response to availability of sufficient remaining processing resources for a second non-neuromorphic performance of the analytical function, assign a portion of the remaining

Abstract: Directed fragmentation of an unmanned aerial vehicle (UAV) is described. In one embodiment, the UAV includes various components, such one or more motors, batteries, sensors, a housing, casing or shell, and a payload for delivery. Additionally, the UAV includes a controller. The controller determines a flight path and controls a flight operation of the UAV. During the flight operation, the controller develops a release timing and a release location for one or more of the components based on the flight path, the flight conditions, and terrain topology information, among other factors. The controller can also detect a disruption in the flight operation of the UAV and, in response, direct fragmentation of one or more of the components apart from the UAV. In that way, a controlled, directed fragmentation of the UAV can be accomplished upon any disruption to the flight operation of the UAV.

Abstract: A computer-implemented method and an apparatus for facilitating staffing of resources receives customer data corresponding to a plurality of customers of an enterprise. At least one intention is predicted for each customer to configure a plurality of intentions. An expected volume of interactions is estimated for at least one time period based on the plurality of intentions. Each interaction in the expected volume of interactions is associated with interaction attributes. Resource data corresponding to a plurality of resources of the enterprise is received. Each resource is associated with a plurality of resource attributes. At least one resource is mapped to each interaction based on a match between resource attributes associated with the at least one resource and the interaction attributes associated with the each interaction. A staffing of the plurality of resources is facilitated based on the mapping of the at least one resource to the each interaction.

Abstract: A system for linking corresponding patient information records is disclosed. A plurality of entities (1,1a) have respective patient databases comprising patient information records (3, 3a). Each entity (1, 1a) has associated therewith a patient identification algorithm (4,4a) for matching corresponding patient information records (3, 3a) of the same patient at different entities (1,1a). A linking subsystem (6) maintains a set of links (7) of a first entity (1) of the plurality of entities (1,1a). The linking subsystem (6) is arranged for linking patient information records (3) of the first entity (1) with corresponding patient information records (3a) of the other entities (1a). A link (ID, RID, RLoc) is established when a given patient information record (3) of the first entity (1) matches a corresponding patient information record (3a) of another entity (1a) based on the patient identification algorithm (4) of the first entity (1).

Abstract: A method for selecting a reduced number of model neurons in a neural network includes generating a first sparse set of non-zero decoding vectors. Each of the decoding vector is associated with a synapse between a first neuron layer and a second neuron layer. The method further includes implementing the neural network only with selected model neurons in the first neuron layer associated with the non-zero decoding vectors.

Abstract: An online system, such as a social networking system, may determine user trends and identify actions to be taken by users that may help optimize revenue for an advertiser. A social networking system may generate offers for an advertiser based on the user trends, user characteristics, and claims about users where the offers include actions determined by the social networking system that users may take to help optimize revenue for advertiser. Venues may also sell ad space, ad inventory, and real-time customer data to advertisers through a social networking system.

Abstract: A method of processing a model of a telecommunications network, the telecommunications network. The method includes dividing the network elements of the telecommunications network into a plurality of different groups, for a first of the group of network elements being modelled, loading computer program code representing the group of network elements being modelled into the computer memory and executing the computer program to model the first group of network elements within a first time period. For a subsequent time period the first group of network elements is represented by the amount of network services consumed on the one or more interfaces with the second group of network elements determined within the first time period, while loading into the computer memory computer program code to model the second group of network elements.

Abstract: A method for customer-based outcome prediction that includes receiving recordings of interactions with customers in a customer group, analyzing the recordings of the interactions to generate interaction data, and building a predictive model using the interaction data, the predictive model identifying a variable relevant to predicting a likelihood of an identified outcome occurring in association with future interactions with customers in the customer group. The method also includes receiving a recording of a current interaction with a first customer, the first customer being in the customer group and analyzing the recording of the current interaction with the first customer to generate current interaction data. Further, the method includes adding the current interaction data to a first customer profile associated with the first customer and applying the predictive model to the first customer profile to predict the likelihood of the identified outcome occurring in association with the current interaction.

Abstract: A system and method for predicting the formulation and processing method and processing parameters for the formation of a biocomposite material is provided. The system and method utilizes the desired properties for the biocomposite material and utilizes these properties m a prediction system to determine the particular formulation, processing method and processing parameters for the formation of a biocomposite material having the desired characteristics. This information is output from the prediction system to a biocomposite material manufacturing system in order to form the biocomposite material and an end product formed therefrom that has the desired characteristics input into the prediction system.

Abstract: A graphical interface module may provide a set of graphical presentations comprising at least: a Likelihood of Delivery chart showing a probability distribution of predicted delivery dates; a Delivery Date Risk Trend chart showing how the completion time for the project predicted according to the Likelihood of Delivery chart has changed over time; and a Burndown chart that shows at least work-items of planned work for the project. Each of the Likelihood of Delivery chart, the Delivery Date Risk Trend chart, and the Burndown chart has a timeline axis.

Abstract: Provided is a process of modifying semantic similarity graphs representative of pair-wise similarity between documents in a corpus, the method comprising obtaining a semantic similarity graph that comprises more than 500 nodes and more than 1000 weighted edges, each node representing a document of a corpus, and each edge weight indicating an amount of similarity between a pair of documents corresponding to the respective nodes connected by the respective edge; obtaining an n-gram indicating that edge weights affected by the n-gram are to be increased or decreased; expanding the n-gram to produce a set of expansion n-grams; adjusting edge weights of edges between pairs of documents in which members of the expanded n-gram set co-occur.

Abstract: An online system using attributes of users to select content for presentation to the users predicts one or more attributes of users whose attributes are unavailable to the online system. For a user with one or more attributes unavailable to the online system, the online system applies a model to attributes of additional users to predict one or more attributes of the user. Attributes of the additional user use in the prediction may include demographic information and interactions with content by the additional users. The online system may determine an accuracy of the model by using the model to predict attributes for users whose attributes are known to the online system and comparing the predicted attributes to the known attributes. If the model's accuracy is less than a threshold value, the online system discontinues using the model to predict attributes of users.

Abstract: Among other things, requests are received from voice assistant devices expressed in accordance with different corresponding protocols of one or more voice assistant frameworks. Each of the requests represents a voiced input by a user to the corresponding voice assistant device. The received requests are re-expressed in accordance with a common request protocol. Based on the received requests, responses to the requests are expressed in accordance with a common response protocol. Each of the responses is re-expressed according to a protocol of the framework with respect to which the corresponding request was expressed. The responses are sent to the voice assistant devices for presentation to the users.

Abstract: Embodiments for monitoring groundwater discharge by one or more processors are described. A groundwater head at each of at least some of a plurality of locations is measured. A groundwater discharge for at least one of the plurality of locations is determined based on the measured groundwater heads. The determined groundwater discharge is compared to a groundwater discharge threshold associated with the at least one of the plurality of locations.

Abstract: A method for assisting game play. The method includes monitoring game play of the user playing a gaming application, wherein the user has a defined task to accomplish, wherein the task is associated with a task type. The method includes determining a task type proficiency rule for the task type based on results of a plurality of players taking on a plurality of tasks having the task type. The method includes determining a player proficiency score for accomplishing the task based on the task type proficiency rule. The method includes determining a user predictive rate of success in accomplishing the task based on the player proficiency score, the task type proficiency rule, and the task. The method includes determining a recommendation for the user based on the user predictive rate of success.

Abstract: A method of transmitting data performed by a server through a network, a method of transmitting data in an ad hoc network, a server, and a data transmitting system are provided. The method of transmitting data performed by a server through a network includes receiving and storing data from a device of a first user; detecting that a device of a second user accesses the server; selecting data from the stored data based on user information; processing the selected data to be optimized for the device of the second user; and transmitting the processed data to the device of the second user, wherein the server is configured to combine data received from a plurality of devices of the first user and transmit the combined data to the device of the second user.

Abstract: A method for providing dispatching services for an on-demand transportation (ODT) service includes determining that a predictive assignment message should be transmitted to a vehicle, generating, in response to the determining that a predictive assignment should be transmitted to a vehicle, the predictive assignment message, and transmitting, to the vehicle, the predictive assignment message. Generating the predictive assignment message uses one or more prediction models computed from historical and real-time ODT service data.

Abstract: Lossless compression of a content item using a neural network trained on content item cohorts. A computing system includes a neural network that is used to train a plurality of symbol prediction models. Each symbol prediction model is trained based on a corresponding cohort of content items. A particular symbol prediction model of the models trained is selected based on an intrinsic characteristic of a particular content item to be losslessly compressed such as, for example, the type or file extension of the content item. The content item is then losslessly compressed based on a set of symbol predictions fed to an arithmetic coder that are generated using the particular symbol prediction model selected.

Abstract: A method for assisting video content searches over a communication network by a user, comprising: A step of determining a first content item (CZ) by said user; A step of automatically extracting semantic data from this first content item; A step of automatically generating queries for at least one service, as a function of semantic data, making it possible to retrieve a set of videos on this service or these services; A step of presenting that set of video content to the user.

Abstract: A method and associated systems for automatic redistribution of virtual machines. A cloud-optimization module selects parameters, such as bandwidth requirements, that characterize an efficiency of a virtual network. It assigns weightings to these parameters based on relative importance of each parameter to the proper operation of the network, where the weightings may be determined as functions of captured network-performance statistics. The module translates the network's topology into a graph in which each node represents a network entity, such as a virtual machine or an application, and each edge represents a connection between two such entities. The module then uses a growing neural gas algorithm to revise the graph and the weightings, and translates the revised graph to a more optimal topology that has redistributed the network entities to operate more efficiently, as measured by the weighted parameters.

Abstract: Simulated network system and method having simulated user profiles and simulated item profiles representing items where the terms used to define characteristics of simulated user profiles and simulated item profiles are based on the same term vocabulary. Simulated sharing and spreading of item profiles by simulated user profiles arranged in the simulated network in a manner that gives a desired level of homophily to the network can produce information that can be presented to a real-world user of a real-world e-commerce system or other user network system, such as a Internet-based social network.

Abstract: Techniques for providing predictive metrics relating to employment positions are provided. A method may include receiving, by a computing device, data relating to a plurality of employment positions, wherein the data is received from a plurality of customers. The computing device may aggregate the data received from the plurality of customers and may determine statistics using the aggregated data, which are based on each of the plurality of employment positions. The computing device may generate one or more predictive metrics relating to the plurality of employment positions using one or more of the statistics.

Abstract: A method for customer-based outcome prediction that includes receiving recordings of interactions with customers in a customer group, analyzing the recordings of the interactions to generate interaction data, and building a predictive model using the interaction data, the predictive model identifying a variable relevant to predicting a likelihood of an identified outcome occurring in association with future interactions with customers in the customer group. The method also includes receiving a recording of a current interaction with a first customer, the first customer being in the customer group and analyzing the recording of the current interaction with the first customer to generate current interaction data. Further, the method includes adding the current interaction data to a first customer profile associated with the first customer and applying the predictive model to the first customer profile to predict the likelihood of the identified outcome occurring in association with the current interaction.

Abstract: A self-aware machine platform is implemented through analyzing operational data of machining tools to achieve machine tool damage assessment, prediction and planning in manufacturing shop floor. Machining processes are first identified by matching similar processes through an ICP algorithm. Machining processes are further clustered by Hotelling's T-squared statistics. Degradation of the machining tool is detected through a trend of the operational data within a cluster of machining processes by a monotonicity test, and the remaining useful life of the machining tool is predicted through a particle filter by extrapolating the trend under a first-order Markov process. In addition, process anomalies across machines are detected through a combination of outlier detection methods including SOMs, multivariate regression, and robust Mahalanobis distance. Warnings and recommendations are flexibly provided to manufacturing shop floor based on policy choice.

Abstract: A computer-implemented system and method for detecting anomalies using sample-based rule identification is provided. Data for data is maintained analytics in a database. A set of anomaly rules is defined. A rare pattern in the data is statistically identified. The identified rare pattern is labeled as at least one of anomaly and non-anomaly based on verification by a domain expert. The set of anomaly rules is adjusted based on the labeled anomaly. Other anomalies in the data are detected and classified by applying the adjusted set of anomaly rules to the data.

Abstract: Certain embodiments involve generating or optimizing a neural network for risk assessment. The neural network can be generated using a relationship between various predictor variables and an outcome (e.g., a condition's presence or absence). The neural network can be used to determine a relationship between each of the predictor variables and a risk indicator. The neural network can be optimized by iteratively adjusting the neural network such that a monotonic relationship exists between each of the predictor variables and the risk indicator. The optimized neural network can be used both for accurately determining risk indicators using predictor variables and determining adverse action codes for the predictor variables, which indicate an effect or an amount of impact that a given predictor variable has on the risk indicator. The neural network can be used to generate adverse action codes upon which consumer behavior can be modified to improve the risk indicator score.

Abstract: A system and method for batched, supervised, in-situ machine learning classifier retraining for malware identification and model heterogeneity.

Abstract: An optical fiber sensor system and a method for determining a location of a disturbance having a signal processor with a plurality of activation cells adapted to react to components of a back-scattered signal and label the disturbance.

Abstract: An apparatus and a method to estimate a degree of user fatigue to video content are disclosed including a feature value calculating unit and a fatigue degree calculating unit. The feature value calculating unit is configured to calculate feature values corresponding to image and sound features in a video content. The fatigue degree calculating unit is configured to calculate a degree of user fatigue to the video content by applying the feature values to a fatigue degree estimation model.

Abstract: Digitized images of the propagated waveforms returned by Time-Domain (TD) sensors reveal the effects of bound water in soil samples, causing errors in the volumetric water content reported for samples due to soil type. Pattern analysis yields not only correct volumetric water content values but also the soil type from which the waveform was taken. Collection and analysis of waveforms from soils having precisely known properties yield computer-generated algorithms to improve the accuracy and number of reporting variables for integrated TD sensors. These algorithms can be deployed in the Cloud for external analysis, or can be incorporated within the sensor to report highly accurate readings of water content, conductivity and soil type under all conditions of these variables with no interpretive burden placed on the user of the sensor. Machine learning algorithms are intended to externally augment data measured by these sensors, or can be added to similar sensors.

Abstract: A first storlet engine is monitored for a first computational algorithm. The storlet engine includes a lookup table including a first characteristic of one or more nodes, a parser unit, and a resource availability monitor. The first computational algorithm is parsed into an abstract syntax tree using a parser unit. A first category of the first computational algorithm is determined by analyzing the abstract syntax tree. A first node of the lookup table with the first characteristic matching the first category for processing the first computational algorithm is identified based on the first category of the computational algorithm. The first computational algorithm is sent to the first node for processing.

Abstract: A neural processing engine may perform processing within a neural processing system and/or artificial neural network. The neural processing engine may be configured to effectively and efficiently perform the type of processing required in implementing a neural processing system and/or an artificial neural network. This configuration may facilitate such processing with neural processing engines having an enhanced computational density and/or processor density with respect to conventional processing units.

Abstract: Methods and apparatus predict non-small cell lung cancer (NSCLC) recurrence using radiomic features extracted from digitized hematoxylin and eosin (H&E) stained slides of a region of tissue demonstrating NSCLC. One example apparatus includes an image acquisition circuit that acquires an image of a region of tissue demonstrating NSCLC, a segmentation circuit that segments a cellular nucleus from the image, a feature extraction circuit that extracts a set of features from the image, a tumor infiltrating lymphocyte (TIL) identification circuit that classifies the segmented nucleus as a TIL or non-TIL, a graphing circuit that constructs a TIL graph and computes a set of TIL graph statistical features, and a classification circuit that computes a probability that the region will experience NSCLC recurrence. The classification circuit may compute a quantitative continuous image-based risk score based on the probability or the image. A treatment plan may be provided based on the risk score.

Abstract: A computer-implemented method includes receiving a transaction, where the transaction includes a plurality of operations and is applicable to a graph database. The transaction is represented by a transaction graph, which is a dependency graph representing dependencies among the plurality of operations of the transaction. The transaction graph is partitioned, by a computer processor, into two or more transaction subgraphs. Each of the two or more transaction subgraphs includes two or more operations of the transaction, and each of the two or more transaction subgraphs is a dependency graph representing dependencies among the two or more operations of the transaction subgraph. The two or more transaction subgraphs are independent of one another. The two or more transaction subgraphs are applied to the graph database in parallel, where applying each transaction subgraph to the graph database includes applying the two or more operations of the transaction subgraph to the graph database.

Abstract: A system and methods store initial states and unsafe states (e.g., safety requirements) of a non-linear model of a control system that interacts with a physical system. The system performs simulations of the non-linear model using a set of sampled initial states, to first generate trajectories (numerical approximations of actual behaviors) over bounded time. The system further determines, for respective pairs of neighboring trajectories: an over-approximation of reachable states as an upper bound of a distance between a pair of neighboring trajectories; a linear over-approximation of the reachable states as piece-wise linear segments over a plurality of time intervals; and whether the linear over-approximation overlaps in any of the piece-wise linear segments with the unsafe states, to verify the non-linear model of the control system is safe.

Abstract: Disclosed herein is a system and methods for determining the alleles, neoantigens, and vaccine composition as determined on the basis of an individual's tumor mutations. Also disclosed are systems and methods for obtaining high quality sequencing data from a tumor. Further, described herein are systems and methods for identifying somatic changes in polymorphic genome data. Finally, described herein are unique cancer vaccines.

Abstract: Methods, apparatus, and other embodiments associated with predicting non-small cell lung cancer (NSCLC) patient response to adjuvant chemotherapy therapy using radiomic features extracted from digitized hematoxylin and eosin (H&E) stained slides of a region of tissue demonstrating NSCLC. One example apparatus includes an image acquisition circuit that acquires an H&E image of a region of tissue demonstrating NSCLC pathology, a segmentation circuit that segments a region of interest (ROI) from the diagnostic radiological image, a feature extraction that extracts a set of discriminative features from the ROI, and a classification circuit that generates a probability that the ROI will experience NSCLC recurrence. The classification circuit may compute a quantitative continuous image-based risk score based on the probability or the image. A prognosis or treatment plan may be provided based on the quantitative continuous image-based risk score.

Abstract: Techniques for automatically suggesting one or more employment positions are provided. A method may include receiving, by a computing device, candidate data relating to an employment position candidate. The computing device may then match the employment position candidate with one or more employment positions based on the received candidate data. Once the employment position candidate is matched with the one or more employment positions, the computing device may generate one or more suggested employment positions for the employment position candidate, wherein the one or more suggested employment positions are automatically generated based on the matching of the employment position candidate with the one or more employment positions.

Abstract: A method can include providing a trained neural network; providing a set of production values where the set includes, for example, a cumulative production value for an interval, an average production value for the interval, a first production value for the interval and a last production value for the interval; and predicting at least one production value for a subsequent interval based at least in part on the trained neural network and the provided set of production values. Various other apparatuses, systems, methods, etc., are also disclosed.

Abstract: There are included a colony evaluation unit 31 that acquires an evaluation result of the cell colony in a cell image obtained by imaging the cell colony, a divided region setting unit 32 that sets a plurality of divided regions by dividing the region of the cell colony according to the evaluation result, a display control unit 34 that displays each of the plurality of divided regions, and a region deformation unit 33 that deforms the divided regions according to a change in the form of the cell colony due to an operation on the cell colony. The display control unit 34 changes a display of the divided regions before the change in form of the cell colony to a display of the divided regions after the deformation.

Abstract: An assertion acceptance value matrix (300) indicates the reliability of assertions, particularly assertions or decisions whether records match or do not match, made by other medical institutions in a federation of medical institutions with different patient record systems and some common patients. Records from different institutions with a high likelihood of matching or not matching are automatically matched or not matched. Those that are ambiguous are manually reviewed. The assertion acceptance value matrix is used to reduce or expedite the manual review.

Abstract: An air-conditioning system control apparatus capable of, even for a room in which a humidity sensor is not installed, accurately estimating an indoor humidity and preventing an indoor temperature and humidity from deviating from comfortable ranges. The air-conditioning system control apparatus includes a thermal characteristic parameter setting unit configured to determine a thermal characteristic parameter of a building thermal model based on a heat equation, a humidity characteristic parameter setting unit configured to determine a humidity characteristic parameter of a building humidity model based on a moisture balance equation in an air-conditioned space, and a control data generation unit configured to generate control data of an air-conditioning apparatus so that a temperature and a humidity of the air-conditioned space are each within a preset range.

Abstract: Systems, methods and devices for evaluating an earth formation intersected by a borehole. Method embodiments may include using a downhole tool in the borehole to generate a plurality of signal pulses responsive to detected radiation; using at least one signal quality characteristic of each pulse of the plurality of pulses to classify each pulse as belonging to at least one subset of a plurality of pulse subsets; and using one of more of the plurality of pulse subsets to estimate a value of a parameter of interest. The signal quality characteristic may be indicative of a pulse pile-up condition being associated with the respective pulse. The signal quality characteristic may include at least one of: i) rising time (slope) of the respective pulse; ii) time between a peak of the respective pulse and a peak of a closest following pulse; iii) pulse width; and iv) skewness of the pulse.

Abstract: A system and method for use in data analysis are provided. The system comprises a data processing utility configured to receive and process input data, comprising: plurality of neural network modules capable for operating in a training mode and in a data processing mode in accordance with the training; a network training utility configured for operating the neural network modules in the training mode utilizing selected set of training data pieces for sequentially training of the neural network modules in a cascade order to reduce an error value with respect to the selected set of the training data pieces for each successive neural network module in the cascade; and an abnormality detection utility configured for sequentially operating said neural network modules for processing input data, and classifying said input data as abnormal upon identifying that all the neural network modules provide error values being above corresponding abnormality detection thresholds.

Abstract: Techniques for providing predictive metrics relating to employment positions are provided. A method may include receiving, by a computing device, data relating to a plurality of employment positions, wherein the data is received from a plurality of customers. The computing device may aggregate the data received from the plurality of customers and may determine statistics using the aggregated data, which are based on each of the plurality of employment positions. The computing device may generate one or more predictive metrics relating to the plurality of employment positions using one or more of the statistics.

Abstract: Systems, methods, and computer-readable storage media that may be used to generate recommendations based on organic search term analysis are provided. One method includes determining conversion path data for a content provider. The method further includes determining a plurality of organic search keywords within the conversion path data. The method further includes analyzing the plurality of organic search keywords within the conversion path data to generate an analysis metric for each of the plurality of organic search keywords. The method further includes selecting one or more of the plurality of organic search keywords based on the analysis metrics for the organic search keywords, and generating one or more recommendations for new content to be published by the content provider based on the selected one or more organic search keywords.

Abstract: A system for predictively provisioning cloud resources based on human behaviors and heuristics includes a computer processor and logic executable by the computer processor. The logic is configured to implement a method. The method includes monitoring a collection of events relating to a customer application as well as monitoring an infrastructure load on resources for the customer application. A causal relationship is evaluated between an event and the infrastructure load. A predictive rule is then constructed based on the causal relationship. Resource requirements are anticipated based on the predictive rule and a provisioning of resources in a service domain is requested for the anticipated resource requirements, according to exemplary embodiments.

Abstract: Computer-implemented methods for identifying or assessing any type of risk and/or opportunity that may arise can include either, alone or in combination, band pass filtering, principal component analysis, random matrix theory analysis, synchronization analysis, and early-warning detection. Each technique can also be viewed as a process that takes a set of inputs and converts it to a set of outputs. These outputs can be used as inputs for a subsequent process or the outputs may be directly actionable for formulating certain economic predictions to make certain decisions.

Abstract: A method includes a computing device receiving data traffic flow and initializing a subroutine to assess alerts for the received data traffic flow; the computer device determining a collection of rules which produce more than a threshold of false positive alerts and more than a threshold of true positive alerts for the assessed alerts; the computer device determining a logical condition and an atomic rule element to add to the collection of rules; and the computer device implementing the added logical condition and atomic rule element if the number of assessed alerts are decreased below a value.

Abstract: Described is a technology by which raw GPS data is processed into segments of a trip, with a predicted mode of transportation (e.g., walking, car, bus, bicycling) determined for each segment. The determined transportation modes may be used to tag the GPS data with transportation mode information, and/or dynamically used. Segments are first characterized as walk segments or non-walk segments based on velocity and/or acceleration. Features corresponding to each of those walk segments or non-walk segments are extracted, and analyzed with an inference model to determine probabilities for the possible modes of transportation for each segment. Post-processing may be used to modify the probabilities based on transitioning considerations with respect to the transportation mode of an adjacent segment. The most probable transportation mode for each segment is selected.