Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Preferences in AI: An overview

Crowdsourcing allows to build hybrid online platforms that combine scalable information systems with the power of human intelligence to complete tasks that are difficult to tackle for current algorithms. Examples include hybrid database systems that use the crowd to fill missing values or to sort items according to subjective dimensions such as picture attractiveness. Current approaches to Crowdsourcing adopt a pull methodology where tasks are published on specialized Web platforms where workers can pick their preferred tasks on a first-come-first-served basis. While this approach has many advantages, such as simplicity and short completion times, it does not guarantee that the task is performed by the most suitable worker. In this paper, we propose and extensively evaluate a different Crowdsourcing approach based on a push methodology. Our proposed system carefully selects which workers should perform a given task based on worker profiles extracted from social networks. Workers and tasks are automatically matched using an underlying categorization structure that exploits entities extracted from the task descriptions on one hand, and categories liked by the user on social platforms on the other hand. We experimentally evaluate our approach on tasks of varying complexity and show that our push methodology consistently yield better results than usual pull strategies.

/pdf/pick-a-crowd-tell-me-what-you-like-and-i-ll-tell-you-what-to-2hn9xae806.pdf

Pick-a-crowd: tell me what you like, and i'll tell you what to do

Group-by and top-k are fundamental constructs in database queries. However, the criteria used for grouping and ordering certain types of data -- such as unlabeled photos clustered by the same person ordered by age -- are difficult to evaluate by machines. In contrast, these tasks are easy for humans to evaluate and are therefore natural candidates for being crowd-sourced.We study the problem of evaluating top-k and group-by queries using the crowd to answer either type or value questions. Given two data elements, the answer to a type question is "yes" if the elements have the same type and therefore belong to the same group or cluster; the answer to a value question orders the two data elements. The assumption here is that there is an underlying ground truth, but that the answers returned by the crowd may sometimes be erroneous. We formalize the problems of top-k and group-by in the crowd-sourced setting, and give efficient algorithms that are guaranteed to achieve good results with high probability. We analyze the crowd-sourced cost of these algorithms in terms of the total number of type and value questions, and show that they are essentially the best possible. We also show that fewer questions are needed when values and types are correlated, or when the error model is one in which the error decreases as the distance between the two elements in the sorted order increases.

/pdf/using-the-crowd-for-top-k-and-group-by-queries-4f1vsv2e3v.pdf

Using the crowd for top-k and group-by queries

Film Theory: An Introduction

By incorporating human workers into the query execution process crowd-enabled databases facilitate intelligent, social capabilities like completing missing data at query time or performing cognitive operators. But despite all their flexibility, crowd-enabled databases still maintain rigid schemas. In this paper, we extend crowd-enabled databases by flexible query-driven schema expansion, allowing the addition of new attributes to the database at query time. However, the number of crowd-sourced mini-tasks to fill in missing values may often be prohibitively large and the resulting data quality is doubtful. Instead of simple crowd-sourcing to obtain all values individually, we leverage the user-generated data found in the Social Web: By exploiting user ratings we build perceptual spaces, i.e., highly-compressed representations of opinions, impressions, and perceptions of large numbers of users. Using few training samples obtained by expert crowd sourcing, we then can extract all missing data automatically from the perceptual space with high quality and at low costs. Extensive experiments show that our approach can boost both performance and quality of crowd-enabled databases, while also providing the flexibility to expand schemas in a query-driven fashion.

Pushing the Boundaries of Crowd-enabled Databases with Query-driven Schema Expansion

By incorporating human workers into the query execution process crowd-enabled databases facilitate intelligent, social capabilities like completing missing data at query time or performing cognitive operators. But despite all their flexibility, crowd-enabled databases still maintain rigid schemas. In this paper, we extend crowd-enabled databases by flexible query-driven schema expansion, allowing the addition of new attributes to the database at query time. However, the number of crowd-sourced mini-tasks to fill in missing values may often be prohibitively large and the resulting data quality is doubtful. Instead of simple crowd-sourcing to obtain all values individually, we leverage the usergenerated data found in the Social Web: By exploiting user ratings we build perceptual spaces, i.e., highly-compressed representations of opinions, impressions, and perceptions of large numbers of users. Using few training samples obtained by expert crowd sourcing, we then can extract all missing data automatically from the perceptual space with high quality and at low costs. Extensive experiments show that our approach can boost both performance and quality of crowd-enabled databases, while also providing the flexibility to expand schemas in a query-driven fashion.

/pdf/pushing-the-boundaries-of-crowd-enabled-databases-with-query-nxybph6r2b.pdf

Pushing the boundaries of crowd-enabled databases with query-driven schema expansion

When issuing user-specific queries, users often have a vaguely defined information need. Skyline queries identify the most "interesting" objects for users' incomplete preferences, which provides users with intuitive query formulation mechanism. However, the applicability of this intuitive query paradigm suffers from a severe drawback. Incomplete preferences on domain values can often lead to impractical skyline result sizes. In particular, this challenge is more critical over categorical domains. This paper addresses this challenge by developing an iterative elicitation framework. While user preferences are collected at each iteration, the framework aims to both minimize user interaction and maximize skyline reduction. The framework allows to identify a reasonably small and focused skyline set, while keeping the query formulation still intuitive for users. All that is needed is answering a few well-chosen questions. We perform extensive experiments to validate the benefits of our strategy and prove that a few questions are enough to acquire a desired manageable skyline set.

/pdf/optimal-preference-elicitation-for-skyline-queries-over-3m2hx2y9ak.pdf

Optimal Preference Elicitation for Skyline Queries over Categorical Domains

The storage, management, and retrieval of entity-related data has always been among the core applications of database systems. However, since nowadays many people access entity collections over the Web (e.g., when searching for products, people, or events), there is a growing need for integrating unconventional types of data into these systems, most notably entity descriptions in unstructured textual form. Prime examples are product reviews, user ratings, tags, and images. While the storage of this data is well-supported by modern database technology, the means for querying it in semantically meaningful ways remain very limited. Consequently, entity-centric search suffers from a growing semantic gap between the users' intended queries and the database's schema. In this paper, we introduce the notion of conceptual views, an innovative extension of traditional database views, which aim to uncover those query-relevant concepts that are primarily reflected by unstructured data. We focus on concepts that are vague in nature and cannot be easily extracted by existing technology (e.g., business phone and romantic movie). After discussing different types of concepts and conceptual queries, we present two case studies, which illustrate how meaningful conceptual information can automatically be extracted from existing data, thus enabling the effective handling of vague real-world query concepts.

/pdf/conceptual-views-for-entity-centric-search-turning-data-into-p75tklmxwo.pdf

Conceptual views for entity-centric search: turning data into meaningful concepts

Performing effective preference-based data retrieval requires detailed and preferentially meaningful structurized information about the current user as well as the items under consideration. A common problem is that representations of items often only consist of mere technical attributes, which do not resemble human perception. This is particularly true for integral items such as movies or songs. It is often claimed that meaningful item features could be extracted from collaborative rating data, which is becoming available through social networking services. However, there is only anecdotal evidence supporting this claim; but if it is true, the extracted information could very valuable for preference-based data retrieval. In this paper, we propose a methodology to systematically check this common claim. We performed a preliminary investigation on a large collection of movie ratings and present initial evidence.

Joachim Selke

Papers

Pushing the Boundaries of Crowd-enabled Databases with Query-driven Schema Expansion

Pushing the boundaries of crowd-enabled databases with query-driven schema expansion

Optimal Preference Elicitation for Skyline Queries over Categorical Domains

Conceptual views for entity-centric search: turning data into meaningful concepts

Extracting Features from Ratings: The Role of Factor Models