We consider the task of learning the parameters of a {\em single} component of a mixture model, for the case when we are given {\em side information} about that component, we call this the "search problem" in mixture models. We would like to solve this with computational and sample complexity lower than solving the overall original problem, where one learns parameters of all components. Our main contributions are the development of a simple but general model for the notion of side information, and a corresponding simple matrix-based algorithm for solving the search problem in this general setting. We then specialize this model and algorithm to four common scenarios: Gaussian mixture models, LDA topic models, subspace clustering, and mixed linear regression. For each one of these we show that if (and only if) the side information is informative, we obtain parameter estimates with greater accuracy, and also improved computation complexity than existing moment based mixture model algorithms (e.g. tensor methods). We also illustrate several natural ways one can obtain such side information, for specific problem instances. Our experiments on real data sets (NY Times, Yelp, BSDS500) further demonstrate the practicality of our algorithms showing significant improvement in runtime and accuracy.

When we first started hearing about machine learning in the early 2010s, it seemed scary at first. Machine learning is essentially using algorithms to calculate trends, value, or other characteristics of specific things based on historical data. Google has even declared itself a machine learning-first company. If you want to learn more about the tactical side of this technology, Eric Enge has a great write-up on Moz explaining how machine learning impacts SEO from a mathematical standpoint. Search engines like to always experiment with how they can use this evolving technology, but here are nine ways we know that they are currently using machine learning and how it relates to SEO or digital marketing.

We are finally going to get the code ready for our sequence 2 sequence neural network which is taking a machine learning approach to keyword extraction. This was part of a larger natural language processing project I did last year.

When it comes to digital marketing, pay per click (PPC) advertising and search engine optimisation (SEO) are arguably two sides of the same coin. However, all too often companies will focus on one at the expense of the other. At ClickThrough Marketing we provide an integrated digital marketing approach. Working with large-scale e-commerce sites, we have learnt the importance of combining PPC and SEO activities to gain greater client and market insight as well as streamline our own internal activities. Here, we look at four ways in which PPC and SEO can work together to deliver better results across the board.

We address the task of generating query suggestions for task-based search. The current state of the art relies heavily on suggestions provided by a major search engine. In this paper, we solve the task without reliance on search engines. Specifically, we focus on the first step of a two-stage pipeline approach, which is dedicated to the generation of query suggestion candidates. We present three methods for generating candidate suggestions and apply them on multiple information sources. Using a purpose-built test collection, we find that these methods are able to generate high-quality suggestion candidates.

There is increasing interest in learning algorithms that involve interaction between human and machine. Comparison-based queries are among the most natural ways to get feedback from humans. A challenge in designing comparison-based interactive learning algorithms is coping with noisy answers. The most common fix is to submit a query several times, but this is not applicable in many situations due to its prohibitive cost and due to the unrealistic assumption of independent noise in different repetitions of the same query. In this paper, we introduce a new weak oracle model, where a non-malicious user responds to a pairwise comparison query only when she is quite sure about the answer. This model is able to mimic the behavior of a human in noise-prone regions. We also consider the application of this weak oracle model to the problem of content search (a variant of the nearest neighbor search problem) through comparisons. More specifically, we aim at devising efficient algorithms to locate a target object in a database equipped with a dissimilarity metric via invocation of the weak comparison oracle. We propose two algorithms termed WORCS-I and WORCS-II (Weak-Oracle Comparison-based Search), which provably locate the target object in a number of comparisons close to the entropy of the target distribution. While WORCS-I provides better theoretical guarantees, WORCS-II is applicable to more technically challenging scenarios where the algorithm has limited access to the ranking dissimilarity between objects. A series of experiments validate the performance of our proposed algorithms.

In this post, we learn about building a basic search engine or document retrieval system using Vector space model. This use case is widely used in information retrieval systems. Given a set of documents and search term(s)/query we need to retrieve relevant documents that are similar to the search query.

Learning sparse linear models with two-way interactions is desirable in many application domains such as genomics. l1-regularised linear models are popular to estimate sparse models, yet standard implementations fail to address specifically the quadratic explosion of candidate two-way interactions in high dimensions, and typically do not scale to genetic data with hundreds of thousands of features. Here we present WHInter, a working set algorithm to solve large l1-regularised problems with two-way interactions for binary design matrices. The novelty of WHInter stems from a new bound to efficiently identify working sets while avoiding to scan all features, and on fast computations inspired from solutions to the maximum inner product search problem. We apply WHInter to simulated and real genetic data and show that it is more scalable and two orders of magnitude faster than the state of the art.

In this paper, we present Agent Assist, a virtual assistant which helps IT support staff to resolve tickets faster. It is essentially a conversation system which provides procedural and often complex answers to queries. This system can ingest knowledge from various sources like application documentation, ticket management systems and knowledge transfer video recordings. It uses an ensemble of techniques like question classification, knowledge graph based disambiguation, information retrieval, etc., to provide quick and relevant solutions to problems from various technical domains and is currently being used in more than 650 projects within IBM.

This paper proposes a simple nearest neighbor search algorithm, which provides the exact solution in terms of the Euclidean distance efficiently. Especially, we present an interesting approach to improve the speed of nearest neighbor search by proper translations of data and query although the task is inherently invariant to the Euclidean transformations. The proposed algorithm aims to eliminate nearest neighbor candidates effectively using their distance lower bounds in nonlinear embedded spaces, and further improves the lower bounds by transforming data and query through product quantized translations. Although our framework is composed of simple operations only, it achieves the state-of-the-art performance compared to existing nearest neighbor search techniques, which is illustrated quantitatively using various large-scale benchmark datasets in different sizes and dimensions.