Nikolaos Nanas

Building and applying a concept hierarchy representation of a user profile

Term dependence is a natural consequence of language use. Its successful representation has been a long standing goal for Information Retrieval research. We present a methodology for the construction of a concept hierarchy that takes into account the three basic dimensions of term dependence. We also introduce a document evaluation function that allows the use of the concept hierarchy as a user profile for Information Filtering. Initial experimental results indicate that this is a promising approach for incorporating term dependence in the way documents are filtered.

Multimodal dynamic optimization: from evolutionary algorithms to artificial immune systems

Multimodal Dynamic Optimisation is a challenging problem, used in this paper as a framework for the qualitative comparison between Evolutionary Algorithms and Artificial Immune Systems. It is argued that while Evolutionary Algorithms have inherent diversity problems that do not allow them to successfully deal with multimodal dynamic optimisation, the biological immune system involves natural processes for maintaining and boosting diversity and thus serves well as a metaphor for tackling this problem. We review the basic evolutionary and immune-inspired approaches to multimodal dynamic optimisation, we identify correspondences and differences and point out essential computation elements.

Autopoiesis, the immune system, and adaptive information filtering

Adaptive information filtering is a challenging and fascinating problem. It requires the adaptation of a representation of a user’s multiple interests to various changes in them. We tackle this dynamic problem with Nootropia, a model inspired by the autopoietic view of the immune system. It is based on a self-organising antibody network that reacts to user feedback in order to define and preserve the user interests. We describe Nootropia in the context of adaptive, content-based document filtering and evaluate it using virtual users. The results demonstrate Nootropia’s ability to adapt to both short-term variations and more radical changes in the user’s interests, and to dynamically control its size and connectivity in the process. Advantages over existing approaches to profile adaptation, such as learning algorithms and evolutionary algorithms are also highlighted.

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.

A review of evolutionary and immune-inspired information filtering

In recent years evolutionary and immune-inspired approaches have been applied to content-based and collaborative filtering. These biologically inspired approaches are well suited to problems like profile adaptation in content-based filtering and rating sparsity in collaborative filtering, due to their distributed and dynamic characteristics. In this paper we introduce the relevant concepts and algorithms and review the state of the art in evolutionary and immune-inspired information filtering. Our intention is to promote the interplay between information filtering and biologically inspired computing and boost developments in this emerging interdisciplinary field.

Personalised news and scientific literature aggregation

Content-based filtering can be deployed for personalised information dissemination on the web, but this is a possibility that has been largely ignored. Nowadays, there are no successful content-based filtering applications available online. Nootropia is an immune-inspired user profiling model for content-based filtering. It has the advantageous property to be able to represent a users multiple interests and adapt to a variety of changes in them. In this paper we describe our early efforts to develop real world personalisation services based on Nootropia. We present, the architecture, implementation, usage and evaluation of the personalised news and paper aggregator, which aggregates news and papers that are relevant to an individuals interests. Our user study shows that Nootropia can effectively learn a users interests and identify relevant information. It also indicates that information filtering is a complicated task with many factors affecting its successful application in a real situation.

Immune-Inspired adaptive information filtering

Adaptive information filtering is a challenging research problem. It requires the adaptation of a representation of a users multiple interests to various changes in them. We investigate the application of an immune-inspired approach to this problem. Nootropia, is a user profiling model that has many properties in common with computational models of the immune system that have been based on Franscisco Varelas work. In this paper we concentrate on Nootropias evaluation. We define an evaluation methodology that uses virtual users to simulate various interest changes. The results show that Nootropia exhibits the desirable adaptive behaviour.

Multi-topic Information Filtering with a Single User Profile

In Information Filtering (IF) a user may be interested in several topics in parallel. But IF systems have been built on representational models derived from Information Retrieval and Text Categorization, which assume independence between terms. The linearity of these models results in user profiles that can only represent one topic of interest. We present a methodology that takes into account term dependencies to construct a single profile representation for multiple topics, in the form of a hierarchical term network. We also introduce a series of non-linear functions for evaluating documents against the profile. Initial experiments produced positive results.

What Happened to Content-Based Information Filtering?

Personalisation can have a significant impact on the way information is disseminated on the web today. Information Filtering can be a significant ingredient towards a personalised web. Collaborative Filtering is already being applied successfully for generating personalised recommendations of music tracks, books, movies and more. The same is not true for Content-Based Filtering. In this paper, we identify some possible reasons for the notable absence of a broad range of personalised information delivery and dissemination services on the web today. We advocate that a more holistic approach to user profiling is required and we discuss the series of still open, challenging research issues raised.

Immune Learning in a Dynamic Information Environment

In Adaptive Information Filtering, the user profile has to be able to define and maintain an accurate representation of the users interests over time. According to Autopoietic Theory, the immune system faces a similar continuous learning problem. It is an organisationally closed network that reacts autonomously to define and preserve the organisms identity. Nootropia is a user profiling model, which has been inspired by this view of the immune system. In this paper, we introduce new improvements to the model and propose a methodology for testing the ability of a user profile to continuously learn a users changing interests in a dynamic information environment. Comparative experiments show that Nootropia outperforms a popular learning algorithm, especially when more than one topic of interest has to be represented.