Typing based communication applications on smartphones, like WhatsApp, can induce emotional exchanges. The effects of an emotion in one session persist across sessions. In this work, we attempt automatic detection by jointly modeling the typing characteristics, and persistence emotion. speed, number mistakes, special characters used, are inferred from Self reports recording states after sessions capture We use data to train a personalized machine learning model for multi-state...

A large population of users gets affected by sudden slowdown or shutdown an enterprise application. System administrators and analysts spend considerable amount time dealing with functional performance bugs. These problems are particularly hard to detect diagnose in most computer systems, since there is a huge system generated supportability data (counters, logs etc.) that need be analyzed. Most often, isn't very clear obvious root cause. Timely identification significant change application...

Network embedding, that aims to learn low-dimensional vector representation of nodes such the network structure is preserved, has gained significant research attention in recent years. However, most state-of-the-art embedding methods are computationally expensive and hence unsuitable for representing billion-scale networks. In this paper, we present LouvainNE, a hierarchical clustering approach embedding. Precisely, employ Louvain, an extremely fast accurate community detection method, build...

Identification of influential users in online social networks allows to facilitate efficient information diffusion a large part the network and thus benefiting diverse applications including viral marketing, disease control, news dissemination. Existing methods have mainly relied on structure only for detection users. In this paper, we enrich approach by proposing fast, efficient, unsupervised algorithm SmartInf detect set identifying anchor nodes from temporal sequence retweets Twitter...

In Affective Computing, different modalities, such as speech, facial expressions, physiological properties, smart-phone usage patterns, and their combinations, are applied to detect the affective states of a user. Keystroke analysis i.e. study typing behavior in desktop computer is found be an effective modality for emotion detection because its reliability, non-intrusiveness low resource overhead. As smartphones proliferate, on smartphone presents equally powerful detection. It has added...

Public transport in suburban cities (covers 80% of the urban landscape) developing regions suffer from lack information Google Transit, unpredictable travel times, chaotic schedules, absence board inside vehicle. Consequently, passengers about exact location where bus is at present as well estimated time to be taken reach desired destination. We find that off-the-shelf deployment existing (non-GPS) localization schemes exhibit high error due sparsity stable and structured outdoor landmarks...

The rate of mental health disorders is rising across the globe. While it significantly affects quality life, an early detection can prevent fatal consequences. Existing literature suggests that mobile based sensing technology be used to determine different conditions like stress, bipolar disorder. In today's smartphone communication, a significant portion on instant messaging apps WhatsApp; thus providing opportunity unobtrusively monitor text input interaction pattern track state. We, in...

WiFi clients must associate to a specific Access Point (AP) communicate over the Internet. Current association methods are based on maximum Received Signal Strength Index (RSSI) implying that client associates strongest AP around it. This is simple scheme has performed well in purely distributed settings. Modern wireless networks, however, increasingly being connected by wired backbone. The backbone allows for out-of-band communication among APs, opening up opportunities improved protocol...

Community detection in single layer, isolated networks has been extensively studied the past decade. However, many real-world systems can be naturally conceptualized as multilayer which embed multiple types of nodes and relations. In this paper, we propose algorithm for detecting communities networks. The crux is based on modularity index Q_M, developed paper. proposed parameter-free, scalable adaptable to complex network structures. More importantly, it simultaneously detect consisting only...

For any initially correlated network after kind of attack where either nodes or edges are removed, we obtain general expressions for the degree-degree probability matrix and degree distribution. We show that proposed analytical approach predicts correct topological changes by comparing evolution assortativity coefficient different strategies intensities in theory simulations. find it is possible to turn an assortative into a disassortative one, vice versa, fine-tuning removal edges....

Event-based online social platforms, such as Meetup and Plancast, have experienced increased popularity rapid growth in recent years. In EBSN setup, selecting suitable venues for hosting events, which can attract a great turnout, is key challenge. this paper, we present deep learning based venue recommendation system DeepVenue provides context driven recommendations the event-hosts to host their events. The crux of proposed model relies on notion similarity between multiple entities venues,...

The development of models for social networks and the spread information therein has become an important field research in recent decades. Here we apply adapt results from theory heterogeneous random graphs to problem modelling Twitter predicting size cascades. We show that a cascade (measured by number users which have retweeted tweet) can be linked largest forward connected component heterogeneous, directed graph with independent edges. discuss different specifications such real compare...

The Experience Sampling Method (ESM) is widely used to collect emotion self-reports train machine learning models for inference. However, as ESM studies are time-consuming and burdensome, participants often withdraw in between. This unplanned withdrawal compels the researchers discard dropout participants' data, significantly impacting quality quantity of self-reports. To address this problem, we leverage only self-reporting similarity across (unlike prior works that apply different...

Passenger comfort is a major factor influencing commuter's decision to avail public transport. Existing studies suggest that factors like overcrowding, jerkiness, traffic congestion etc. correlate well passenger's (dis)comfort. An online survey conducted with more than 300 participants from 12 different countries reveals personalized and context dependent influence passenger during travel by Leveraging on these findings, we identify correlations between level dynamic parameters, implement...

Smartphones provide the capability to perform in-situ sampling of human behavior using Experience Sampling Method (ESM). Designing an ESM schedule involves probing user repeatedly at suitable moments collect self-reports. Timely probe generation high fidelity responses while keeping rate low is challenging. In mobile-based ESM, timeliness also impacted by user's availability respond self-report request. Thus, a good design must consider - <italic...

This paper explores the feasibility of automatically extracting passwords from a user's daily activity logs, such as her Facebook activity, phone etc. As an example, smartphone might ask user: "Today morning whom did you receive SMS?" In this paper, we observe that infrequent activities (i.e., outliers) can be memorable and unpredictable. Building on observation, have developed end to system ActivPass experimented with 70 users. With logs Facebook, browsing history, call SMSs, achieves 95%...

The growth in the market for cab companies like Uber has opened door to high-income options drivers. However, order boost their income, drivers many a time resort accepting trips which increases stress resulting poor driving quality and accidents serious cases. Every driver handles differently trip recommendation thus needs be on personalized level. In this paper, we explore historical data compute its impact various behavioral features, captured through vehicle-mounted GPS inertial sensors....

In this paper, we develop an analytical framework to measure the vulnerability of superpeer networks against attack. Two different kinds attacks namely deterministic and degree dependent attack have been introduced here. We formally model with help bimodal structure graph dynamics. Our analysis shows that fraction superpeers their connectivity profound impact upon stability network. The results obtained from theoretical are validated through simulation. agreement between simulation...