Users of online search engines often find it difficult to express their need for information in the form of a query. However, if the user can identify examples of the kind of documents they require then they can employ a technique known as relevance feedback. Relevance feedback covers a range of techniques intended to improve a user's query and facilitate retrieval of information relevant to a user's information need. In this paper we survey relevance feedback techniques. We study both automatic techniques, in which the system modifies the user's query, and interactive techniques, in which the user has control over query modification. We also consider specific interfaces to relevance feedback systems and characteristics of searchers that can affect the use and success of relevance feedback systems.

No abstract

Machine-learned models are o en described as "black boxes". In many real-world applications however, models may have to sacrice predictive power in favour of human-interpretability. When this is the case, feature engineering becomes a crucial task, which requires signi cant and time-consuming human e ort. Whilst some features are inherently static, representing properties that cannot be in uenced (e.g., the age of an individual), others capture characteristics that could be adjusted (e.g., the daily amount of carbohydrates taken). Nonetheless, once a model is learned from the data, each prediction it makes on new instances is irreversible -assuming every instance to be a static point located in the chosen feature space.ere are many circumstances however where it is important to understand (i) why a model outputs a certain prediction on a given instance, (ii) which adjustable features of that instance should be modi ed, and nally (iii) how to alter such a prediction when the mutated instance is input back to the model.In this paper, we present a technique that exploits the internals of a tree-based ensemble classi er to o er recommendations for transforming true negative instances into positively predicted ones. We demonstrate the validity of our approach using an online advertising application. First, we design a Random Forest classi er that e ectively separates between two types of ads: low (negative) and high (positive) quality ads (instances). en, we introduce an algorithm that provides recommendations that aim to transform a low quality ad (negative instance) into a high quality one (positive instance). Finally, we evaluate our approach on a subset of the active inventory of a large ad network, Yahoo Gemini.

WHY IS IT IMPORTANT TO ENGAGE USERS?o In today's wired world, users have enhanced expectations about their interactions with technology … resulting in increased competition amongst the purveyors and designers of interactive systems.o In addition to utilitarian factors, such as usability, we must consider the hedonic and experiential factors of interacting with technology, such as fun, fulfillment, play, and user engagement. WHAT IS USER ENGAGEMENT (UE)? (I)o "The state of mind that we must attain in order to enjoy a representation of an action" so that we may experience computer worlds "directly, without mediation or distraction" (Laurel, 1993, pp. 112-113, 116).o "Engagement is a user's response to an interaction that gains maintains, and encourages their attention, particularly when they are intrinsically motivated" (Jacques, 1996, p. 103).o A quality of user experience that depends on the aesthetic appeal, novelty, and usability of the system, the ability of the user to attend to and become involved in the experience, and the user's overall evaluation of the experience. Engagement depends on the depth of participation the user is able to achieve with respect to each experiential attribute (O'Brien & Toms, 2008).o "…explain[s] how and why applications attract people to use them" (Sutcliffe, 2010, p. 3). 8 WHAT IS UE? (II)o User engagement is a quality of the user experience that emphasizes the positive aspects of interaction -in particular the fact of being captivated by the technology (Attfield et al, 2011 14 SOME CAVEATS (I)o This tutorial assumes that web application are "properly designed"• We do not look into how to design good web site (although some user engagement measurement may inform for an enhanced design).o This tutorial is based on "published research" literature • We do not know how each individual company and organization measure user engagement (although we guess some common baselines).o This tutorial focuses on web applications that users "chose" to engage with• A web tool that has to be used e.g. for work purpose, is totally different (users have no choice).o This tutorial is not an "exhaustive" account of all existing works • We focus on work that we came across and that has influenced us; if we have missed something important, let us know. 15 SOME CAVEATS (II)o This tutorial focuses on web applications that are widely used by "anybody" on a "large-scale"• User engagement in the game industry or education have different characteristics.o This tutorial does not focus on the effect of advertisements on user engagement• We assume that web applications that display ads do so in a "normal" way so that to not annoy or frustrate users.o This tutorial looks at user engagement at web application "level"• Although we use examples and may refer to specific sites or types of applications, we do not focus on any particular applications.o This tutorial is not about "how" to influence user engagement  16

No abstract