Learning and Recognizing the Places We Go

Echo-ID: Smartphone Placement Region Identification for Context-Aware Computing

Region-function combinations are essential for smartphones to be intelligent and context-aware. The prerequisite for providing intelligent services is that the device can recognize the contextual region in which it resides. The existing region recognition schemes are mainly based on indoor positioning, which require pre-installed infrastructures or tedious calibration efforts or memory burden of precise locations. In addition, location classification recognition methods are limited by either their recognition granularity being too large (room-level) or too small (centimeter-level, requiring training data collection at multiple positions within the region), which constrains the applications of providing contextual awareness services based on region function combinations. In this paper, we propose a novel mobile system, called Echo-ID, that enables a phone to identify the region in which it resides without requiring any additional sensors or pre-installed infrastructure. Echo-ID applies Frequency Modulated Continuous Wave (FMCW) acoustic signals as its sensing medium which is transmitted and received by the speaker and microphones already available in common smartphones. The spatial relationships among the surrounding objects and the smartphone are extracted with a signal processing procedure. We further design a deep learning model to achieve accurate region identification, which calculate finer features inside the spatial relations, robust to phone placement uncertainty and environmental variation. Echo-ID requires users only to put their phone at two orthogonal angles for 8.5 s each inside a target region before use. We implement Echo-ID on the Android platform and evaluate it with Xiaomi 12 Pro and Honor-10 smartphones. Our experiments demonstrate that Echo-ID achieves an average accuracy of 94.6% for identifying five typical regions, with an improvement of 35.5% compared to EchoTag. The results confirm Echo-ID's robustness and effectiveness for region identification.

A survey on predicting personal mobility

Purpose

Location‐prediction enables the next generation of location‐based applications. The purpose of this paper is to provide a historical summary of research in personal location‐prediction. Location‐prediction began as a tool for network management, predicting the load on particular cellular towers or WiFi access points. With the increasing popularity of mobile devices, location‐prediction turned personal, predicting individuals' next locations given their current locations.

Design/methodology/approach

This paper includes an overview of prediction techniques and reviews several location‐prediction projects comparing the raw location data, feature extraction, choice of prediction algorithms and their results.

Findings

A new trend has emerged, that of employing additional context to improve or expand predictions. Incorporating temporal information enables location‐predictions farther out into the future. Appending place types or place names can improve predictions or develop prediction applications that could be used in any locale. Finally, the authors explore research into diverse types of context, such as people's personal contacts or health activities.

Originality/value

This overview provides a broad background for future research in prediction.

Applying Commonsense Reasoning to Place Identification

Recent mobile computing applications try to automatically identify the places visited by the user from a log of GPS readings. Such applications reverse geocode the GPS data to discover the actual places (shops, restaurants, etc.) where the user has been. Unfortunately, because of GPS errors, the actual addresses and businesses being visited cannot be extracted unambiguously and often only a list of candidate places can be obtained. Commonsense reasoning can notably help the disambiguation process by invalidating some unlikely findings (e.g., a user visiting a cinema in the morning). This paper illustrates the use of Cyc—an artificial intelligence system comprising a database of commonsense knowledge—to improve automatic place identification. Cyc allows to probabilistically rank the list of candidate places in consideration of the commonsense likelihood of that place being actually visited on the basis of the user profile, the time of the day, what happened before, and so forth. The system has been evaluated using real data collected from a mobile computing application.

Discovering routines from large-scale human locations using probabilistic topic models

In this work, we discover the daily location-driven routines that are contained in a massive real-life human dataset collected by mobile phones. Our goal is the discovery and analysis of human routines that characterize both individual and group behaviors in terms of location patterns. We develop an unsupervised methodology based on two differing probabilistic topic models and apply them to the daily life of 97 mobile phone users over a 16-month period to achieve these goals. Topic models are probabilistic generative models for documents that identify the latent structure that underlies a set of words. Routines dominating the entire group's activities, identified with a methodology based on the Latent Dirichlet Allocation topic model, include “going to work late”, “going home early”, “working nonstop” and “having no reception (phone off)” at different times over varying time-intervals. We also detect routines which are characteristic of users, with a methodology based on the Author-Topic model. With the routines discovered, and the two methods of characterizing days and users, we can then perform various tasks. We use the routines discovered to determine behavioral patterns of users and groups of users. For example, we can find individuals that display specific daily routines, such as “going to work early” or “turning off the mobile (or having no reception) in the evenings”. We are also able to characterize daily patterns by determining the topic structure of days in addition to determining whether certain routines occur dominantly on weekends or weekdays. Furthermore, the routines discovered can be used to rank users or find subgroups of users who display certain routines. We can also characterize users based on their entropy. We compare our method to one based on clustering using K-means. Finally, we analyze an individual's routines over time to determine regions with high variations, which may correspond to specific events.

Place recognition and automatic semantic annotation via the Whereabouts diary

Purpose

Modern handheld devices provided with localization capabilities can create a diary of the user whereabouts, and provide a description of the user habits and a complement of the user profile in several applications. The places we go, in fact, reveal something about us; for example, two persons can be matched as compatible given the fact that they visit the same places. The purpose of this paper is to describe the Whereabouts diary in this context.

Design/methodology/approach

This paper presents the Whereabouts diary, an application/service to log the places visited by the user and to label them, in an automatic way, with descriptive semantic information. Web‐retrieved information, and the temporal patterns with which different places are visited, can be used to automatically define meaningful semantic labels to the visited places.

Findings

The paper verified that such diary application can be created and can effectively classify the places visited by the user. In particular, geocoding and white‐pages web services were used to extract information about a place, and Bayesian networks to classify places on the basis of the time at which they have been visited.

Research limitations/implications

The paper discusses this implementation, and presents experimental results. Experiments show that the identification of places and the accuracy of the place classification mechanism are effective, while the accuracy of geocoding and white‐pages retrieval should be improved.

Originality/value

This paper shows the novel Whereabouts diary application. Several mechanisms presented are original to this approach. In addition, several applications that can exploit the diary are illustrated.

Discovering personally meaningful places

The discovery of a person's meaningful places involves obtaining the physical locations and their labels for a person's places that matter to his daily life and routines. This problem is driven by the requirements from emerging location-aware applications, which allow a user to pose queries and obtain information in reference to places, for example, “home”, “work” or “Northwest Health Club”. It is a challenge to map from physical locations to personally meaningful places due to a lack of understanding of what constitutes the real users' personally meaningful places. Previous work has explored algorithms to discover personal places from location data. However, we know of no systematic empirical evaluations of these algorithms, leaving designers of location-aware applications in the dark about their choices.

Our work remedies this situation. We extended a clustering algorithm to discover places. We also defined a set of essential evaluation metrics and an interactive evaluation framework. We then conducted a large-scale experiment that collected real users' location data and personally meaningful places, and illustrated the utility of our evaluation framework. Our results establish a baseline that future work can measure itself against. They also demonstrate that that our algorithm discovers places with reasonable accuracy and outperforms the well-known K-Means clustering algorithm for place discovery. Finally, we provide evidence that shapes more complex than “points” are required to represent the full range of people's everyday places.