M-Dyaa Albakour

On sparsity and drift for effective real-time filtering in microblogs

In this paper, we approach the problem of real-time filtering in the Twitter Microblogging platform. We adapt an effective traditional news filtering technique, which uses a text classifier inspired by Rocchios relevance feedback algorithm, to build and dynamically update a profile of the users interests in real-time. In our adaptation, we tackle two challenges that are particularly prevalent in Twitter: sparsity and drift. In particular, sparsity stems from the brevity of tweets, while drift occurs as events related to the topic develop or the interests of the user change. First, to tackle the acute sparsity problem, we apply query expansion to derive terms or related tweets for a richer initialisation of the user interests within the profile. Second, to deal with drift, we modify the user profile to balance between the importance of the short-term interests, i.e. emerging subtopics, and the long-term interests in the overall topic. Moreover, we investigate an event detection method from Twitter and newswire streams to predict times at which drift may happen. Through experiments using the TREC Microblog track 2012, we show that our approach is effective for a number of common filtering metrics such as the users utility, and that it compares favourably with state-of-the-art news filtering baselines. Our results also uncover the impact of different factors on handling topic drifting.

Deriving query suggestions for site search

Modern search engines have been moving away from simplistic interfaces that aimed at satisfying a users need with a single‐shot query. Interactive features are now integral parts of web search engines. However, generating good query modification suggestions remains a challenging issue. Query log analysis is one of the major strands of work in this direction. Although much research has been performed on query logs collected on the web as a whole, query log analysis to enhance search on smaller and more focused collections has attracted less attention, despite its increasing practical importance. In this article, we report on a systematic study of different query modification methods applied to a substantial query log collected on a local website that already uses an interactive search engine. We conducted experiments in which we asked users to assess the relevance of potential query modification suggestions that have been constructed using a range of log analysis methods and different baseline approaches. The experimental results demonstrate the usefulness of log analysis to extract query modification suggestions. Furthermore, our experiments demonstrate that a more fine‐grained approach than grouping search requests into sessions allows for extraction of better refinement terms from query log files.

Exploring ant colony optimisation for adaptive interactive search

Search engines have become much more interactive in recent years which has triggered a lot of work in automatically acquiring knowledge structures that can assist a user in navigating through a document collection. Query log analysis has emerged as one of the most promising research areas to automatically derive such structures. We explore a biologically inspired model based on ant colony optimisation applied to query logs as an adaptive learning process that addresses the problem of deriving query suggestions. A user interaction with the search engine is treated as an individual ants journey and over time the collective journeys of all ants result in strengthening more popular paths which leads to a corresponding term association graph that is used to provide query modification suggestions. This association graph is being updated in a continuous learning cycle. In this paper we use a novel automatic evaluation framework based on actual query logs to explore the effect of different parameters in the ant colony optimisation algorithm on the performance of the resulting adaptive query suggestion model. We also use the framework to compare the ant colony approach against a state-of-the-art baseline. The experiments were conducted with query logs collected on a university search engine over a period of several years.

On the Importance of Venue-Dependent Features for Learning to Rank Contextual Suggestions

Suggesting venues to a user in a given geographic context is an emerging task that is currently attracting a lot of attention. Existing studies in the literature consist of approaches that rank candidate venues based on different features of the venues and the user, which either focus on modeling the preferences of the user or the quality of the venue. However, while providing insightful results and conclusions, none of these studies have explored the relative effectiveness of these different features. In this paper, we explore a variety of user-dependent and venue-dependent features and apply state-of-the-art learning to rank approaches to the problem of contextual suggestion in order to find what makes a venue relevant for a given context. Using the test collection of the TREC 2013 Contextual Suggestion track, we perform a number of experiments to evaluate our approach. Our results suggest that a learning to rank technique can significantly outperform a Language Modelling baseline that models the positive and negative preferences of the user. Moreover, despite the fact that the contextual suggestion task is a personalisation task (i.e. providing the user with personalised suggestions of venues), we surprisingly find that user-dependent features are less effective than venue-dependent features for estimating the relevance of a suggestion.

Sentence-Level attachment prediction

Attachment prediction is the task of automatically identifying email messages that should contain an attachment. This can be useful to tackle the problem of sending out emails but forgetting to include the relevant attachment (something that happens all too often). A common Information Retrieval (IR) approach in analyzing documents such as emails is to treat the entire document as a bag of words. Here we propose a finer-grained analysis to address the problem. We aim at identifying individual sentences within an email that refer to an attachment. If we detect any such sentence, we predict that the email should have an attachment. Using part of the Enron corpus for evaluation we find that our finer-grained approach outperforms previously reported document-level attachment prediction in similar evaluation settings. A second contribution this paper makes is to give another successful example of the ‘wisdom of the crowd’ when collecting annotations needed to train the attachment prediction algorithm. The aggregated non-expert judgements collected on Amazon’s Mechanical Turk can be used as a substitute for much more costly expert judgements.

Moving towards adaptive search in digital libraries

Search applications have become very popular over the last two decades, one of the main drivers being the advent of the Web. Nevertheless, searching on the Web is very different to searching on smaller, often more structured collections such as digital libraries, local Web sites, and intranets. One way of helping the searcher locating the right information for a specific information need in such a collection is by providing well-structured domain knowledge to assist query modification and navigation. There are two main challenges which we will both address in this chapter: acquiring the domain knowledge and adapting it automatically to the specific interests of the user community. We will outline how in digital libraries a domain model can automatically be acquired using search engine query logs and how it can be continuously updated using methods resembling ant colony behaviour.

AutoEval: An Evaluation Methodology for Evaluating Query Suggestions Using Query Logs

User evaluations of search engines are expensive and not easy to replicate. The problem is even more pronounced when assessing adaptive search systems, for example system-generated query modification suggestions that can be derived from past user interactions with a search engine. Automatically predicting the performance of different modification suggestion models before getting the users involved is therefore highly desirable. AutoEval is an evaluation methodology that assesses the quality of query modifications generated by a model using the query logs of past user interactions with the system. We present experimental results of applying this methodology to different adaptive algorithms which suggest that the predicted quality of different algorithms is in line with user assessments. This makes AutoEval a suitable evaluation framework for adaptive interactive search engines

Diversifying contextual suggestions from location-based social networks

In this paper, we study the emerging Information Retrieval (IR) task of contextual suggestion in location-based social networks. The aim of this task is to make personalised recommendations of venues for entertainments or activities whilst visiting a city, by appropriately representing the context of the user, such as their location and personal interests. Instead of only representing the specific low-level interests of a user, our approach is driven by estimates of the high-level categories of venues that the user may be interested in. Moreover, we argue that an effective model for contextual suggestion should not only promote the categories that the user is interested in, but it should also be capable of eliminating redundancy by diversifying the recommended venues in the sense that they should cover various categories of interest to the given user. Therefore, we adapt web search result diversification approaches to the task of contextual suggestion. For categorising the venues, we use the category classifications employed by location-based social networks such as FourSquare, urban guides such as Yelp, and a large collection of web pages, the ClueWeb12 corpus, to build a textual classifier that is capable of predicting the category distribution for a certain venue given its web page. We thoroughly evaluate our approach using the TREC 2013 Contextual Suggestion track. We conduct a number of experiments where we consider venues from the closed environments of both FourSquare and Yelp, and the general web using the ClueWeb12 corpus. Our empirical results suggest that category diversification consistently improves the effectiveness of the recommendation model over a reasonable baseline that only considers the similarity between the users profile and venue. The results also give insights on the effectiveness of our approach with different types of users.

Profile-Based Summarisation for Web Site Navigation

Information systems that utilise contextual information have the potential of helping a user identify relevant information more quickly and more accurately than systems that work the same for all users and contexts. Contextual information comes in a variety of types, often derived from records of past interactions between a user and the information system. It can be individual or group based. We are focusing on the latter, harnessing the search behaviour of cohorts of users, turning it into a domain model that can then be used to assist other users of the same cohort. More specifically, we aim to explore how such a domain model is best utilised for profile-biased summarisation of documents in a navigation scenario in which such summaries can be displayed as hover text as a user moves the mouse over a link. The main motivation is to help a user find relevant documents more quickly. Given the fact that the Web in general has been studied extensively already, we focus our attention on Web sites and similar document collections. Such collections can be notoriously difficult to search or explore. The process of acquiring the domain model is not a research interest here; we simply adopt a biologically inspired method that resembles the idea of ant colony optimisation. This has been shown to work well in a variety of application areas. The model can be built in a continuous learning cycle that exploits search patterns as recorded in typical query log files. Our research explores different summarisation techniques, some of which use the domain model and some that do not. We perform task-based evaluations of these different techniques—thus of the impact of the domain model and profile-biased summarisation—in the context of Web site navigation.

Using virtual documents to move information retrieval and knowledge management closer together

While Information Retrieval approaches typically rely on a bag-of-word approach and are therefore fairly shallow, Knowledge Management is based on a deep semantic representation. Both allow users to sift through huge volumes of information and to identify the relevant document or answer for a particular information need. Obviously, these areas go beyond this simple information access scenario. Nevertheless, it is also a fact that these areas form fairly separate communities. We propose an idea of how to make use of techniques coming from both ends of the spectrum and combine them in methods that are more powerful than each of them individually. Our idea is based on turning structured information into virtual documents built to preserve the concepts and relations inherent in the semantically rich data before we apply Information Retrieval methods.