Search

Inferring user profiles based on texts created by users on social networks has a variety of applications in recommender systems such as job offering, item recommendation, and targeted advertisement. The problem becomes more challenging when working with short texts like tweets on Twitter, or posts on Facebook. This work aims at proposing an integrated framework based on Dempster–Shafer theory of evidence, word embedding, and k -means clustering for user profiling problem, which is capable of not only working well with short texts but also dealing with uncertainty inherently in user texts. The proposed framework is essentially composed of three phases: (1) Learning abstract concepts at multiple levels of abstraction from user corpora; (2) Evidential inference and combination for user modeling; and (3) User profile extraction. Particularly, in the first phase, a word embedding technique is used to convert preprocessed texts into vectors which capture semantics of words in user corpus, and then k -means clustering is utilized for learning abstract concepts at multiple levels of abstraction, each of which reflects appropriate semantics of user profiles. In the second phase, by considering each document in user corpus as an evidential source that carries some partial information for inferring user profiles, we first infer a mass function associated with each user document by maximum a posterior estimation, and then apply Dempster’s rule of combination for fusing all documents’ mass functions into an overall one for the user corpus. Finally, in the third phase, we apply the so-called pignistic probability principle to extract top- n keywords from user’s overall mass function to define the user profile. Thanks to the ability of combining pieces of information from many documents, the proposed framework is flexible enough to be scaled when input data coming from not only multiple modes but different sources on web environments. Besides, the resulting profiles are interpretable, visualizable, and compatible in practical applications. The effectiveness of the proposed framework is validated by experimental studies conducted on datasets crawled from Twitter and Facebook.

This paper proposes a new outlier detection approach that measures the degree of outlierness for each instance in a given dataset. The proposed model utilizes a mass-based dissimilarity measure to address the weaknesses of neighbor-based outlier models while detecting local outliers in the dataset within a variety of data point densities. In particular, it first applies a hierarchical partitioning technique to generate a set of tree-like nested structure partitions for the input dataset, and then a mass-based dissimilarity measure is defined to quantify the dissimilarity between two data instances given the generated hierarchical partition structure. After that, for each data instance, a context set is obtained by gathering the neighbors around it with the $k$ lowest mass dissimilarities, and based on those context sets, a mass-based local outlier score model is introduced to compute the outlierness for each individual instance. The proposed approach fundamentally changes the perspective of the outlier model by using the mass-based measurement instead of the distance-based functions used in most neighbor-based methods. A comprehensive experiment conducted on both synthetic and real-world datasets demonstrates that the proposed approach is not only competitive with the existing state-of-the-art outlier detection models but is also an efficient and effective alternative for local outlier detection methods.

Short texts-based user profiling is a challenging task despite its benefits to many applications such as job offering, item recommendation, and targeted advertisement. This work proposes an evidential reasoning framework for user profiling problem, which is capable of not only working with short texts but also dealing with uncertainty inherently in user corpora. The framework consists of three phases: representation learning, inference and combination, and user profile extraction. Particularly, in the first phase a word embedding technique is used to convert words into semantic vectors, then k-means clustering is utilized for learning abstract concepts that hide inside user plain texts, and then organizing the learned concepts into a hierarchical structure. In the second phase, by considering each document in user corpus as an evidential source that partially contributing to user profiles, and taking the concept hierarchy into account, we infer a mass function associated with each user document by maximum a posterior estimation, then apply Dempster’s rule of combination for fusing all documents’ mass functions into an overall one for the user corpus. In the third phase, we apply the so-called pignistic probability principle to extract top-n keywords from user’s overall mass function to define the user profile. Thanks to the ability of combining pieces of information from many documents, the proposed framework is flexible enough to be scaled when input data coming from not only multiple modes but different sources on web environments. Empirical results on datasets crawled from Twitter and Facebook validate the effectiveness of our proposed framework.

Background Patient journals have been used as valuable resources in clinical studies. However, the full potential value of such journals can be undermined by inefficiencies and ambiguities associated with handwritten patient reports. The increasing number of mobile phones and mobile-based health care approaches presents an opportunity to improve communications from patients to clinicians and clinical researchers through the use of digital patient journals. Objective The objective of this project was to develop a smartphone-based platform that would enable patients to record events and symptoms on the same timeline as clinical data collected by wearable sensors. Methods This platform consists of two major components: a smartphone for patients to record their journals and wireless sensors for clinical data collection. The clinical data and patient records are then exported to a clinical researcher interface, and the data and journal are processed and combined into a single time-series graph for analysis. This paper gives a block diagram of the platform's principal components and compares its features to those of other methods but does not explicitly discuss the process of design or development of the system. Results As a proof of concept, body temperature data were obtained in a 4-hour span from a 22-year-old male, during which the subject simultaneously recorded relevant activities and events using the iPhone platform. After export to a clinical researcher's desktop, the digital records and temperature data were processed and fused into a single time-series graph. The events were filtered based on specific keywords to facilitate data analysis. Conclusions We have developed a user-friendly patient journal platform, based on widely available smartphone technology, that gives clinicians and researchers a simple method to track and analyze patient activities and record the activities on a shared timeline with clinical data from wearable devices.