1. Natural Language Processing: models of parsing and sentiment analysis.
2. Methodology: techniques and tools to aid in the development of scientific (mostly cognitive) models and software.

My undergraduate studies were in Mathematics and Computation, which led to an MSc and then PhD in Computer Science. I held a Post-Doctoral position related to Cognitive Science before joining the University of Hertfordshire as a lecturer in Computer Science.

I have supervised/assisted a wide range of projects, but below are some notes on those I particularly led or had input into.

Machine Learning

Most of my published work falls in the area of "machine learning" - a way of making computers useful by building up algorithms or data-structures from data.

> Natural Language Processing

Sentiment Analysis

In a KTP project, I worked with the associate and company to develop some text classifiers, exploring the impact of feature selection and imbalanced data.

• P.C.R. Lane, D. Clarke, and P. Hender, 'On developing robust models for favourability analysis: Model choice, feature sets and imbalanced data,' Decision Support Systems, 53:712-18, 2012.  https://doi.org/10.1016/j.dss.2012.05.028 - http://hdl.handle.net/2299/9017

Simple Synchrony Networks

During my PhD, I worked with Dr. James Henderson to create the Simple Synchrony Network, a form of recurrent neural network specialised for parsing natural language.

• P.C.R. Lane and J.B. Henderson, 'Incremental syntactic parsing of natural language corpora with Simple Synchrony Networks,' IEEE Transactions on Knowledge and Data Engineering, 13:219-231, 2001.  https://doi.org/10.1109/69.917562 - http://hdl.handle.net/2299/8618
• P.C.R. Lane and J.B. Henderson, 'Towards effective parsing with neural networks: Inherent generalisations and bounded resource effects,' Applied Intelligence, 19:83-100, 2003.  https://doi.org/10.1023/A:1023820807862

> Cognitive Science

Cognitive science is an attractive application area for machine learning. Firstly, in modelling the human learning process, we construct models using simulated data and compare their behaviours with those of humans. Secondly, in the methodology of constructing such models, we can use machine-learning techniques as ways to generate or improve models.

CHREST

I was introduced to CHREST in my post-doctoral position, where I worked with Prof. Fernand Gobet and Prof. Peter Cheng on CHREST-AVOW - a version of CHREST specialised to learn from diagrams of electric circuits. 

CHREST is worth studying from a machine-learning perspective because it builds up a form of discrimination network from information which it obtains through a guided attention system.  I have created a fairly complete implementation of CHREST and developed some small models:

• P.C.R. Lane and F. Gobet, 'CHREST models of implicit learning and board game interpretation', in J.Bach, B.Goertzel and M.Ikle (Eds.), Proceedings of the Fifth Conference on Artificial General Intelligence. Lecture Notes in Computer Science(), vol 7716, pp.  148-157, 2012. (Springer, Berlin, Heidelberg) https://doi.org/10.1007/978-3-642-35506-6_16 (open access)
• P.C.R. Lane and F. Gobet, 'Using chunks to categorise chess positions', in M.Bramer and M.Petridis (Eds.) Research and Development in Intelligent Systems XXIX. SGAI 2012, pp. 93-106, 2012. (Springer London) https://doi.org/10.1007/978-1-4471-4739-8_7

Methodology for Developing Cognitive Models

Prof. Fernand Gobet and I have developed some ideas for improving the methodology of developing cognitive models. These ideas are partly drawn from machine learning and partly from software engineering: how can we create better cognitive models? Firstly, all theoretically important behaviour in the implementation should be documented and tested. Secondly, parameters in models need to be properly optimised. These ideas are presented in:

• P.C.R. Lane and F. Gobet, 'A theory-driven testing methodology for developing scientific software', Journal of Experimental and Theoretical Artificial Intelligence, 24:421-56, 2012. https://doi.org/10.1080/0952813X.2012.695443
• P.C.R. Lane and F. Gobet, 'Evolving non-dominated parameter sets for computational models from multiple experiments', Journal of Artificial General Intelligence, 4:1-30, 2013. https://doi.org/10.2478/jagi-2013-0001 (open access) - code

An extension of the above ideas involves automating the actual construction process of cognitive models using a program-synthesis technique, such as Genetic Programming. The most recent version of this is the project Genetically Evolving Models in Science - the website links to all relevant publications, but my own selection includes:

• P.C.R. Lane, L.K. Bartlett, N. Javed, A. Pirrone and F. Gobet, 'Evolving understandable cognitive models', in T.C.Steward (Ed.), Proceedings of the 20^th International Conference on Cognitive Modelling, pp.176-82, 2022. (University Park, PA: Applied Cognitive Science Lab, Penn State.) pdf - Proceedings
• P.C.R. Lane, P.D. Sozou, F. Gobet and M. Addis, 'Analysing psychological data by evolving computational models', in A. Wilhelm and H. Kestler (Eds.), Analysis of Large and Complex Data. Studies in Classification, Data Analysis, and Knowledge Organization, pp.587-597, 2016. (Springer, Cham.) https://doi.org/10.1007/978-3-319-25226-1_50

> Cloud Computing

I worked with Dr. Na Helian and others in understanding how to use genetic algorithms to optimise scheduling architectures:

• P.C.R. Lane, N. Helian, M.H. Bodla, M. Zheng and P.M. Moggridge, 'Dynamic hierarchical structure optimisation for cloud computing job scheduling', in J.L. Jiménez Laredo, J.I. Hidalgo, K.O. Babaagba (Eds.), Applications of Evolutionary Computation. EvoApplications 2022. Lecture Notes in Computer Science(), vol 13224, pp.301-316, 2022. (Springer, Cham.) https://doi.org/10.1007/978-3-031-02462-7_20

Mathematics

Mathematical principles underpin much of computer science, and ideas from discrete mathematics, linear algebra, logic, topology and so on will be found throughout my teaching and research.

One overtly mathematical piece of research work was completed with the late Dr. Andreas Albrecht, related to search landscapes for k-SAT instances:

• A.A. Albrecht, P.C.R. Lane and K. Steinhöfel, 'Analysis of local search landscapes for k-SAT instances,' Mathematics in Computer Science, 3:465-488, 2010. https://doi.org/10.1007/s11786-010-0040-7 - http://hdl.handle.net/2299/4632