Tanel Tammet

A Resolution Theorem Prover for Intuitonistic Logic

We use the general scheme of building resolution calculi (also called the inverse method) originating from S.Maslov and G.Mints to design and implement a resolution theorem prover for intuitionistic logic. A number of search strategies are introduced and proved complete. The resolution method is shown to be a decision procedure for a new syntactically described decidable class of intuitionistic logic. The performance of our prover is compared with the performance of a tableau prover for intuitionistic logic presented in [12], using both the benchmarks from the latter and the theorems from J. von Plato-s constructive geometry [9].

Proof strategies in linear logic

Linear logic, introduced by J.-Y. Girard, is a refinement of classical logic providing means for controlling the allocation of “resources”. It has aroused considerable interest from both proof theorists and computer scientists. In this paper we investigate methods for automated theorem proving in propositional linear logic. Both the “bottom-up” (tableaux) and “top-down” (resolution) proof strategies are analyzed. Various modifications of sequent rules and efficient search strategies are presented along with the experiments performed with the implemented theorem provers.

Towards Efficient Subsumption

We propose several methods for writing efficient subsumption procedures for non-unit clauses, tested in practice as parts incorporated into the Gandalf family of theorem provers. Versions of Gandalf exist for classical logic, first order intuitionistic logic and type theory.

Personalized Context-Aware Recommendations in SMARTMUSEUM: Combining Semantics with Statistics

Our goal is to enhance on-site personalized access and recommendations for cultural heritage. We have designed and implemented the SMARTMUSEUM platform using adaptive and privacy preserving user profiling. The described recommendation system relies on combining a semantics/ontologies based approach with a data mining/statistics based approach. The paper presents the architecture and main methods of the system.

Smart City: A Rule-based Tourist Recommendation System

We are presenting the architecture of a personalised recommendation system using probabilistic reasoning. The system creates plans for visiting interesting objects and events on the trip, based on the personal preference profile. We will provide a short description of the system architecture, its main components and the probabilistic reasoner along with a small ruleset example with score calculations. The described architecture and components comprise the first iteration of the Smart City system.

Resolution, Inverse Method and the Sequent Calculus

We discuss the general scheme of building resolution calculi (also called the inverse method) originating from S. Maslov and G. Mints. A survey of resolution calculi for various nonclassical logic is presented, along with several common properties these calculi possess.

RFID-based Communications for a Self-Organising Robot Swarm

We investigate the practical questions of building a self-organizing robot swarm, using the iRobot Roomba cleaning robot as an experimental platform. Our goal is to employ self-organization for enhancing the cleaning efficiency of a Roomba swarm. The implementation uses RFID tags both for object and location-based task recognition as well as graffiti- or stigmata-style communication between robots.Easily modifiable rule systems are used for object ontologies and automatic task generation. Long-term planning and central coordination are avoided.

Condensed Detachment is Complete for Relevance Logic: A Computer-Aided Proof

The condensed detachment ruleD is a combination of modus ponens with a minimal amount of substitution. EarlierD has been shown to be complete for intuitionistic and classical implicational logic but incomplete forBCK andBCI logic. We show thatD is complete for the relevance logic. One of the main steps is the proof of the formula ((a →a) →a) →a found in interaction with our resolution theorem prover. Various strategies of generating consequences of the axioms and choosing best ones for the next iteration were tried until the proof was found.

Sightsmap: Crowd-Sourced Popularity of the World Places

We analyse and combine a number of world-wide crowd-sourced geotagged databases with the goal to locate, describe and rate potential tourism targets in any area in the world. In particular, we address the problem of finding representative names and top POIs for popular areas, with the main focus on sightseeing. The results are demonstrated on the sightsmap.com site presenting a zoomable and pannable tourism popularity heat map along with popularity-sorted POI markers for concrete objects.

A survey on socio-semantic information retrieval

Abstract The rise of the Social Web and advances in the Semantic Web provides unprecedented possibilities for the development of novel methods to enhance the information retrieval (IR) process by including varying degrees of semantics. We shed light on the corresponding notion of semantically-enhanced information retrieval by presenting state-of-the art techniques in related research areas. We describe techniques based on the main processes of a typical IR workflow and map them onto three main types of semantics, which vary from formal semantic knowledge representations and content-based semantics to social semantics emerging through usage and user interactions.