We propose a novel network-based approach for location estimation in social media that integrates evidence of the social tie strength between users for improved location estimation. Concretely, we propose a location estimator – FriendlyLocation– that leverages the relationship between the strength of the tie between a pair of users, and the distance between the pair. Based on an examination of over 100 million geo-encoded tweets and 73 million Twitter user profiles, we identify several factors such as the number of followers and how the users interact that can strongly reveal the distance between a pair of users. We use these factors to train a decision tree to distinguish between pairs of users who are likely to live nearby and pairs of users who are likely to live in different areas. We use the results of this decision tree as the input to a maximum likelihood estimator to predict a user’s location. We find that this proposed method significantly improves the results of location estimation relative to a state-of-the-art technique. Our system reduces the average error distance for 80% of Twitter users from 40 miles to 21 miles using only information from the user’s friends and friends-of-friends, which has great significance for augmenting traditional social media and enriching location-based services with more refined and accurate location estimates.

In the United States, smartphone ownership surpassed 69.5 million in February 2011 with a large portion of those users (20%) downloading applications (apps) that enhance the usability of a device by adding additional functionality. a large percentage of apps are written specifically to utilize the geographical position of a mobile device. One of the prime factors in developing location prediction models is the use of historical data to train such a model. with larger sets of training data, prediction algorithms become more accurate; however, the use of historical data can quickly become a downfall if the GPS stream is not collected or processed correctly. Inaccurate or incomplete or even improperly interpreted historical data can lead to the inability to develop accurately performing prediction algorithms. As GPS chipsets become the standard in the ever increasing number of mobile devices, the opportunity for the collection of GPS data increases remarkably. the goal of this study is to build a comprehensive system that addresses the following challenges: (1) collection of GPS data streams in a manner such that the data is highly usable and has a reduction in errors; (2) processing and reduction of the collected data in order to prepare it and make it highly usable for the creation of prediction algorithms; (3) creation of prediction/labeling algorithms at such a level that they are viable for commercial use. This study identifies the key research problems toward building the CaPPture (collection, processing, prediction) system.

No description available.

Since August 2010, Facebook has entered the self-reported positioning world by providing the check-in service to its users. This service allows users to share their physical location using the GPS receiver in their mobile devices such as a smart-phone, tablet, or smart-watch. Over the years, big datasets of recorded check-ins have been collected with increasing popularity of social networks. Analyzing the check-in datasets reveals valuable information and patterns in users’ check-in behavior as well as places check-in history. The analysis results can be used in several areas including business planning and financial decisions, for instance providing location-based deals. In this thesis, we leverage novel data mining methodology to learn from big check-in data and predict the next check-in place based on only places’ history and with no reference to individual users. To this end, we study a large Facebook check-in dataset. This dataset has a high level of noise in location coordinates due to multiple collection sources, which are users’ mobile devices. The research question is how we can leverage a noise impact reduction technique to enhance performance of prediction model. We design our own noise handling mechanism to deal with feature noise. The predictive model is generated by Random Forest classification algorithm in a shared-memory parallel environment. We represent how the performance of predictors is enhanced by minimizing noise impacts. The solution is a preprocessing feature noise cleansing approach implemented in R and works fast for big check-in datasets.

In a telecommunications network locating user equipment (paging) is a common procedure. Proposed functionality for 4G and 5G allows for eNB initiated paging via X2 interfaces. In this thesis machine learning algorithms were evaluated in order to reduce page signalling. Additionally, two paging schemes based on machine learning were proposed and compared to a common method of paging through cost models. The results show that signalling cost can be reduced by up to 80%.

Since some of the location of where the users posted their tweets collected by social media company have varied accuracy, and some are missing. We want to use those tweets with highest accuracy to help fill in the data of those tweets with incomplete information. To test our algorithm, we used the sets of social media data from a city, we separated them into training sets, where we know all the information, and the testing sets, where we intentionally pretend to not know the location. One prediction method that was used in (Dukler, Han and Wang, 2016) requires appending one-hot encoding of the location to the bag of words matrix to do Location Oriented Nonnegative Matrix Factorization (LONMF). We improve further on this algorithm by introducing iterative LONMF. We found that when the threshold and number of iterations are chosen correctly, we can predict tweets location with higher accuracy than using LONMF.

abstract: Time series forecasting is the prediction of future data after analyzing the past data for temporal trends. This work investigates two fields of time series forecasting in the form of Stock Data Prediction and the Opioid Incident Prediction. In this thesis, the Stock Data Prediction Problem investigates methods which could predict the trends in the NYSE and NASDAQ stock markets for ten different companies, nine of which are part of the Dow Jones Industrial Average (DJIA). A novel deep learning model which uses a Generative Adversarial Network (GAN) is used to predict future data and the results are compared with the existing regression techniques like Linear, Huber, and Ridge regression and neural network models such as Long-Short Term Memory (LSTMs) models. In this thesis, the Opioid Incident Prediction Problem investigates methods which could predict the location of future opioid overdose incidences using the past opioid overdose incidences data. A similar deep learning model is used to predict the location of the future overdose incidences given the two datasets of the past incidences (Connecticut and Cincinnati Opioid incidence datasets) and compared with the existing neural network models such as Convolution LSTMs, Attention-based Convolution LSTMs, and Encoder-Decoder frameworks. Experimental results on the above-mentioned datasets for both the problems show the superiority of the proposed architectures over the standard statistical models. / Dissertation/Thesis / Masters Thesis Computer Science 2019

This work has two objectives: a) to predict people's future locations, and b) to predict when they will be at given locations. Current location-based applications react to the user's current location. The progression from location-awareness to location-prediction can enable the next generation of proactive, context-predicting applications. Existing location-prediction algorithms predict someone's next location. In contrast, this dissertation predicts someone's future locations. Existing algorithms use a sequence of locations and predict the next location in the sequence. This dissertation incorporates temporal information as timestamps in order to predict someone's location at any time in the future. Sequence predictors based on Markov models have been shown to be effective predictors of someone's next location. This dissertation applies a Markov model to two-dimensional, timestamped location information to predict future locations. This dissertation also predicts when someone will be at a given location. These predictions can support presence or understanding co-workers’ routines. Predicting the times that someone is going to be at a given location is a very different and more difficult problem than predicting where someone will be at a given time. A location-prediction application may predict one or two key locations for a given time, while there could be hundreds of correct predictions for times of the day that someone will be in a given location. The approach used in this dissertation, a heuristic model loosely based on Market Basket Analysis, is the first to predict when someone will arrive at any given location. The models are applied to sparse, WiFi mobility data collected on PDAs given to 275 college freshmen. The location-prediction model predicts future locations with 78-91% accuracy. The temporal-prediction model achieves 33-39% accuracy. If a tolerance of plus/minus twenty minutes is allowed, the prediction rates rise to 77%-91%. This dissertation shows the characteristics of the timestamped, location data which lead to the highest number of correct predictions. The best data cover large portions of the day, with less than three locations for any given timestamp. / Ph. D.

Doctor of Philosophy / Department of Computing and Information Sciences / David Gustafson / We use probabilistic graphs to predict the location of swarms over 100 steps in simulations in grid worlds. One graph can be used to make predictions for worlds of different dimensions. The worlds are constructed from a single 5x5 square pattern, each square of which may be either unoccupied or occupied by an obstacle or a target. Simulated robots move through the worlds avoiding the obstacles and tagging the targets. The interactions between the robots and the robots and the environment lead to behavior that, even in deterministic simulations, can be difficult to anticipate. The graphs capture the local rate and direction of swarm movement through the pattern. The graphs are used to create a transition matrix, which along with an occupancy matrix, can be used to predict the occupancy in the patterns in the 100 steps using 100 matrix multiplications. In the future, the graphs could be used to predict the movement of physical swarms though patterned environments such as city blocks in applications such as disaster response search and rescue. The predictions could assist in the design and deployment of such swarms and help rule out undesirable behavior.

In mobile computing systems, predicting the next location of a mobile wireless user has gained interest over the past decade. Location prediction may have a wide-range of application areas such as network load balancing, advertising and web page prefetching. In the literature, there exist many location predictors which are divided into two main classes: domain-independent and domain-specific. Song et al. compare the prediction accuracy of the domain-independent predictors from four major families, namely, Markov-based, compression-based, PPM and SPM predictors on Dartmouth&#039 / s campus-wide Wi-Fi mobility data. As a result, the low-order Markov predictors are found as the best predictor. In another work, Bayir et al. propose a domain-specific location predictor (LPMP) as the application of a framework used for discovering mobile cell phone user profiles. In this thesis, we evaluate LPMP and the best Markov predictor with Dartmouth&#039 / s campus-wide Wi-Fi mobility data in terms of accuracy. We also propose a simple method which improves the accuracy of LPMP slightly in the location prediction part of LPMP. Our results show that the accuracy of the best Markov predictor is better than that of LPMP in total. However, interestingly, LPMP yields more accurate results than the best Markov predictor does for the users with the low prediction accuracy.