Aditya Telang

One Size Does Not Fit All: Toward User- and Query-Dependent Ranking for Web Databases

With the emergence of the deep web, searching web databases in domains such as vehicles, real estate, etc., has become a routine task. One of the problems in this context is ranking the results of a user query. Earlier approaches for addressing this problem have used frequencies of database values, query logs, and user profiles. A common thread in most of these approaches is that ranking is done in a user- and/or query-independent manner. This paper proposes a novel query- and user-dependent approach for ranking query results in web databases. We present a ranking model, based on two complementary notions of user and query similarity, to derive a ranking function for a given user query. This function is acquired from a sparse workload comprising of several such ranking functions derived for various user-query pairs. The model is based on the intuition that similar users display comparable ranking preferences over the results of similar queries. We define these similarities formally in alternative ways and discuss their effectiveness analytically and experimentally over two distinct web databases.

Indexing and matching trajectories under inconsistent sampling rates

Quantifying the similarity between two trajectories is a fundamental operation in analysis of spatio-temporal databases. While a number of distance functions exist, the recent shift in the dynamics of the trajectory generation procedure violates one of their core assumptions; a consistent and uniform sampling rate. In this paper, we formulate a robust distance function called Edit Distance with Projections (EDwP) to match trajectories under inconsistent and variable sampling rates through dynamic interpolation. This is achieved by deploying the idea of projections that goes beyond matching only the sampled points while aligning trajectories. To enable efficient trajectory retrievals using EDwP, we design an index structure called TrajTree. TrajTree derives its pruning power by employing the unique combination of bounding boxes with Lipschitz embedding. Extensive experiments on real trajectory databases demonstrate EDwP to be up to 5 times more accurate than the state-of-the-art distance functions. Additionally, TrajTree increases the efficiency of trajectory retrievals by up to an order of magnitude over existing techniques.

Detecting localized homogeneous anomalies over spatio-temporal data

The last decade has witnessed an unprecedented growth in availability of data having spatio-temporal characteristics. Given the scale and richness of such data, finding spatio-temporal patterns that demonstrate significantly different behavior from their neighbors could be of interest for various application scenarios such as—weather modeling, analyzing spread of disease outbreaks, monitoring traffic congestions, and so on. In this paper, we propose an automated approach of exploring and discovering such anomalous patterns irrespective of the underlying domain from which the data is recovered. Our approach differs significantly from traditional methods of spatial outlier detection, and employs two phases—(i) discovering homogeneous regions, and (ii) evaluating these regions as anomalies based on their statistical difference from a generalized neighborhood. We evaluate the quality of our approach and distinguish it from existing techniques via an extensive experimental evaluation.

A Graph-Based Approach for Multi-folder Email Classification

This paper presents a novel framework for multi-folder email classification using graph mining as the underlying technique. Although several techniques exist (e.g., SVM, TF-IDF, n-gram) for addressing this problem in a delimited context, they heavily rely on extracting high-frequency keywords, thus ignoring the inherent structural aspects of an email (or document in general) which can play a critical role in classification. Some of the models (e.g., n-gram) consider only the words without taking into consideration where in the structure these words appear together. This paper presents a supervised learning model that leverages graph mining techniques for multi-folder email classification. A ranking formula is presented for ordering the representative - common and recurring - substructures generated from pre-classified emails. These ranked representative substructures are then used for categorizing incoming emails. This approach is based on a global ranking model that incorporates several relevant parameters for email classification and overcomes numerous problems faced by extant approaches used for multi-folder classification. A number of parameters which influence the generation of representative substructures are analyzed, reexamined, and adapted to multiple folders. The effect of graph representations has been analyzed. The effectiveness of the proposed approach has been validated experimentally.

Trajectory aware macro-cell planning for mobile users

We handle the problem of efficient user-mobility driven macro-cell planning in cellular networks. As cellular networks embrace heterogeneous technologies (including long range 3G/4G and short range WiFi, Femto-cells, etc.), most traffic generated by static users gets absorbed by the short-range technologies, thereby increasingly leaving mobile user traffic to macro-cells. To this end, we consider a novel approach that factors in the trajectories of mobile users as well as the impact of city geographies and their associated road networks for macro-cell planning. Given a budget k of base-stations that can be upgraded, our approach selects a deployment that improves the most number of user trajectories. The generic formulation incorporates the notion of quality of service of a user trajectory as a parameter to allow different application-specific requirements, and operator choices. We show that the proposed trajectory utility maximization problem is NP-hard, and design multiple heuristics. We evaluate our algorithms with real and synthetic datasets emulating different city geographies to demonstrate their efficacy. For instance, with an upgrade budget k of 20%, our algorithms perform 3-8 times better in improving the user quality of service on trajectories when compared to greedy location-based base-station upgrades.

People in motion: Spatio-temporal analytics on Call Detail Records

The data about how people move in a city can be potentially used by various enterprises and government organizations to strategically optimize their operations and maximize their revenue. However, fine-grained and real-time data is currently unavailable to the enterprises. We believe that Cellular Network operators can deliver such data and insights to enterprises. Call records collected in the networks embed a wealth of information about where, when and how a large fraction of the city moves. However, this information is untapped; a majority of the cellular operators are not deriving spatio-temporal insights or monetizing the data that is already available. In this paper, we demonstrate “People in Motion”: an end-to-end Hadoop-based system with a library of spatio-temporal algorithms that operates on the call record data to derive business insights. We identify the hangouts and trajectories of users with different interests. Finally, we demonstrate a visual analytics tool that facilitates business users to compute, compare and contrast the importance of spatial regions at different times for different categories of users.

Query-By-Keywords (QBK): Query Formulation Using Semantics and Feedback

The staples of information retrieval have been querying and search , respectively, for structured and unstructured repositories. Processing queries over known, structured repositories (e.g., Databases) has been well-understood, and search has become ubiquitous when it comes to unstructured repositories (e.g., Web). Furthermore, searching structured repositories has been explored to a limited extent. However, there is not much work in querying unstructured sources. We argue that querying unstructured sources is the next step in performing focused retrievals. This paper proposed a new approach to generate queries from search-like inputs for unstructured repositories. Instead of burdening the user with schema details, we believe that pre-discovered semantic information in the form of taxonomies, relationship of keywords based on context, and attribute & operator compatibility can be used to generate query skeletons. Furthermore, progressive feedback from users can be used to improve the accuracy of query skeletons generated.