Umut Ozertem

Learning to suggest: a machine learning framework for ranking query suggestions

We consider the task of suggesting related queries to users after they issue their initial query to a web search engine. We propose a machine learning approach to learn the probability that a user may find a follow-up query both useful and relevant, given his initial query. Our approach is based on a machine learning model which enables us to generalize to queries that have never occurred in the logs as well. The model is trained on co-occurrences mined from the search logs, with novel utility and relevance models, and the machine learning step is done without any labeled data by human judges. The learning step allows us to generalize from the past observations and generate query suggestions that are beyond the past co-occurred queries. This brings significant gains in coverage while yielding modest gains in relevance. Both offline (based on human judges) and online (based on millions of user interactions) evaluations demonstrate that our approach significantly outperforms strong baselines.

A Study of Medication-Taking and Unobtrusive, Intelligent Reminding

Poor medication adherence is one of the major causes of illness and of treatment failure in the United States. The objective of this study was to conduct an initial evaluation of a context-aware reminder system, which generated reminders at an opportune time to take the medication. Ten participants aged 65 or older, living alone and managing their own medications, participated in the study. Participants took a low-dose vitamin C tablet twice daily at times that they specified. Participants were considered adherent if they took the vitamin within 90 minutes (before or after) of the prescribed time. Adherence and activity in the home was measured using a system of sensors, including an instrumented pillbox. There were three phases of the study: baseline, in which there was no prompting; time-based, in which there was prompting at the prescribed times for pill-taking; and context-aware, in which participants were only prompted if they forgot to take their pills and were likely able to take their pills. The context-based prompting resulted in significantly better adherence (92.3%) as compared to time-based (73.5%) or no prompting (68.1%) conditions (p < 0.0002, chi(2) = 17.0). In addition, subjects had better adherence in the morning than in the evening. We have shown in this study that a system that generates reminders at an opportune time to take the medication significantly improves adherence. This study indicates that context-aware prompting may provide improved adherence over standard time-based reminders.

Mean shift spectral clustering

In recent years there has been a growing interest in clustering methods stemming from the spectral decomposition of the data affinity matrix, which are shown to present good results on a wide variety of situations. However, a complete theoretical understanding of these methods in terms of data distributions is not yet well understood. In this paper, we propose a spectral clustering based mode merging method for mean shift as a theoretically well-founded approach that enables a probabilistic interpretation of affinity based clustering through kernel density estimation. This connection also allows principled kernel optimization and enables the use of anisotropic variable-size kernels to match local data structures. We demonstrate the proposed algorithms performance on image segmentation applications and compare its clustering results with the well-known Mean Shift and Normalized Cut algorithms.

Automatic Online Evaluation of Intelligent Assistants

Voice-activated intelligent assistants, such as Siri, Google Now, and Cortana, are prevalent on mobile devices. However, it is challenging to evaluate them due to the varied and evolving number of tasks supported, e.g., voice command, web search, and chat. Since each task may have its own procedure and a unique form of correct answers, it is expensive to evaluate each task individually. This paper is the first attempt to solve this challenge. We develop consistent and automatic approaches that can evaluate different tasks in voice-activated intelligent assistants. We use implicit feedback from users to predict whether users are satisfied with the intelligent assistant as well as its components, i.e., speech recognition and intent classification. Using this approach, we can potentially evaluate and compare different tasks within and across intelligent assistants ac-cording to the predicted user satisfaction rates. Our approach is characterized by an automatic scheme of categorizing user-system interaction into task-independent dialog actions, e.g., the user is commanding, selecting, or confirming an action. We use the action sequence in a session to predict user satisfaction and the quality of speech recognition and intent classification. We also incorporate other features to further improve our approach, including features derived from previous work on web search satisfaction prediction, and those utilizing acoustic characteristics of voice requests. We evaluate our approach using data collected from a user study. Results show our approach can accurately identify satisfactory and unsatisfactory sessions.

Continuous Activity Monitoring and Intelligent Contextual Prompting to Improve Medication Adherence

Poor medication adherence is a serious medical problem, particularly in older adults. Various solutions have been developed to remind people to take their medications, but these systems are usually simple time-based alarm systems that are not particularly effective. We describe a system that is context aware, and that utilizes information about past patterns of behavior plus the current context to provide prompts at the appropriate time and place. A case study from our initial deployment of the system to eleven older adults illustrates the possibilities and advantages of context aware prompting systems.

Editorial IEEE Transactions on Neural Networks and Learning Systems 2016 and Beyond

“Happy New Year!” At the beginning of 2016, I would like to take this opportunity to wish everyone a very happy, healthy, and prosperous new year! It is my great honor and privilege to serve as the Editor-in-Chief (EiC) of the IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (TNNLS), and I am excited to write this Editorial to start a new journey with you all.

Behavior-driven clustering of queries into topics

Categorization of web-search queries in semantically coherent topics is a crucial task to understand the interest trends of search engine users and, therefore, to provide more intelligent personalization services. Query clustering usually relies on lexical and clickthrough data, while the information originating from the user actions in submitting their queries is currently neglected. In particular, the intent that drives users to submit their requests is an important element for meaningful aggregation of queries. We propose a new intent-centric notion of topical query clusters and we define a query clustering technique that differs from existing algorithms in both methodology and nature of the resulting clusters. Our method extracts topics from the query log by merging missions, i.e., activity fragments that express a coherent user intent, on the basis of their topical affinity. Our approach works in a bottom-up way, without any a-priori knowledge of topical categorization, and produces good quality topics compared to state-of-the-art clustering techniques. It can also summarize topically-coherent missions that occur far away from each other, thus enabling a more compact user profiling on a topical basis. Furthermore, such a topical user profiling discriminates the stream of activity of a particular user from the activity of others, with a potential to predict future user search activity.

Self-Consistent Locally Defined Principal Surfaces

Principal curves and surfaces play an important role in dimensionality reduction applications of machine learning and signal processing. Vaguely defined, principal curves are smooth curves that pass through the middle of the data distribution. This intuitive definition is ill posed and to this day researchers have struggled with its practical implications. Two main causes of these difficulties are: (i) the desire to build a self-consistent definition using global statistics (for instance conditional expectations), and (ii) not decoupling the definition of the principal curve from the data samples. In this paper, we introduce the concept of principal sets, which are the union of all principal surfaces with a particular dimensionality. The proposed definition of principal surfaces provides rigorous conditions for a point to satisfy that can be evaluated using only the gradient and Hessian of the probability density at the point of interest. Since the definition is decoupled from the data samples, any density estimator could be employed to obtain a probability distribution expression and identify the principal surfaces of the data under this particular model.

Spectral feature projections that maximize Shannon mutual information with class labels

Determining optimal subspace projections that can maintain task-relevant information in the data is an important problem in machine learning and pattern recognition. In this paper, we propose a nonparametric nonlinear subspace projection technique that maintains class separability maximally under the Shannon mutual information (MI) criterion. Employing kernel density estimates for nonparametric estimation of MI makes possible an interesting marriage of kernel density estimation-based information theoretic methods and kernel machines, which have the ability to determine nonparametric nonlinear solutions for difficult problems in machine learning. Significant computational savings are achieved by translating the definition of the desired projection into the kernel-induced feature space, which leads to obtain analytical solution.

Mobile query reformulations

Users frequently interact with web search systems on their mobile devices via multiple modalities, including touch and speech. These interaction modes are substantially different from the user experience on desktop search. As a result, system designers have new challenges and questions around understanding the intent on these platforms. In this paper, we study the query reformulation patterns in mobile logs. We group query reformulations based on their input method into four categories; text-text, text-voice, voice-text and voice-voice. We discuss the unique characteristics of each of these groups by comparing them against each other and desktop logs. We also compare the distribution of reformulation types (e.g. adding/dropping words) against desktop logs and show that there are new classes of reformulations that are caused by errors in speech recognition. Our results suggest that users do not tend to switch between different input types (e.g. voice and text). Voice to text switches are largely caused by speech recognition errors, and text to voice switches are unlikely to be about the same intent.