Searcharxiv⌕ Search

arXiv subjects

Dave Cliff

Publications and source records attributed to Dave Cliff.

24 records · Page 2Linked to original sources
arXiv2018

Using Stock Prices as Ground Truth in Sentiment Analysis to Generate Profitable Trading Signals

The increasing availability of "big" (large volume) social media data has motivated a great deal of research in applying sentiment analysis to predict the movement of prices within financial markets. Previous work in this field investigates how the true sentiment of text (i.e. positive or negative opinions) can be used for financial predictions, based on the assumption that sentiments expressed online are representative of the true market sentiment. Here we consider the converse idea, that using the stock price as the ground-truth in the system may be a better indication of sentiment. Tweets are labelled as Buy or Sell dependent on whether the stock price discussed rose or fell over the following hour, and from this, stock-specific dictionaries are built for individual companies. A Bayesian classifier is used to generate stock predictions, which are input to an automated trading algorithm. Placing 468 trades over a 1 month period yields a return rate of 5.18%, which annualises to approximately 83% per annum. This approach performs significantly better than random chance and outperforms two baseline sentiment analysis methods tested.

cs.CE↗
arXiv2018

BSE: A Minimal Simulation of a Limit-Order-Book Stock Exchange

This paper describes the design, implementation, and successful use of the Bristol Stock Exchange (BSE), a novel minimal simulation of a centralised financial market, based on a Limit Order Book (LOB) such as is common in major stock exchanges. Construction of BSE was motivated by the fact that most of the world's major financial markets have automated, with trading activity that previously was the responsibility of human traders now being performed by high-speed autonomous automated trading systems. Research aimed at understanding the dynamics of this new style of financial market is hampered by the fact that no operational real-world exchange is ever likely to allow experimental probing of that market while it is open and running live, forcing researchers to work primarily from time-series of past trading data. Similarly, university-level education of the engineers who can create next-generation automated trading systems requires that they have hands-on learning experience in a sufficiently realistic teaching environment. BSE as described here addresses both those needs: it has been successfully used for teaching and research in a leading UK university since 2012, and the BSE program code is freely available as open-source on GitHuB.

cs.CE↗
arXiv2011

Large-scale Complex IT Systems

This paper explores the issues around the construction of large-scale complex systems which are built as 'systems of systems' and suggests that there are fundamental reasons, derived from the inherent complexity in these systems, why our current software engineering methods and techniques cannot be scaled up to cope with the engineering challenges of constructing such systems. It then goes on to propose a research and education agenda for software engineering that identifies the major challenges and issues in the development of large-scale complex, software-intensive systems. Central to this is the notion that we cannot separate software from the socio-technical environment in which it is used.

cs.SE↗
arXiv2011

Hybrid complex network topologies are preferred for component-subscription in large-scale data-centres

We report on experiments exploring the interplay between the topology of the complex network of dependent components in a large-scale data-centre, and the robustness and scaling properties of that data-centre. In a previous paper [1] we used the SPECI large-scale data-centre simulator [2] to compare the robustness and scaling characteristics of data-centres whose dependent components are connected via Strogatz-Watts small-world (SW) networks [3], versus those organized as Barabasi-Albert scale-free (SF) networks [4], and found significant differences. In this paper, we present results from using the Klemm-Eguiliz (KE) construction method [5] to generate complex network topologies for data-centre component dependencies. The KE model has a control parameter μ\in[0,1]\inR that determines whether the networks generated are SW (0<μ<<1) or SF (μ=1) or a "hybrid" network topology part-way between SW and SF (0<μ<1). We find that the best scores for system-level performance metrics of the simulated data-centres are given by "hybrid" values of μ significantly different from pure-SW or pure-SF.

cs.DC↗
arXiv2010

Effects of component-subscription network topology on large-scale data centre performance scaling

Modern large-scale date centres, such as those used for cloud computing service provision, are becoming ever-larger as the operators of those data centres seek to maximise the benefits from economies of scale. With these increases in size comes a growth in system complexity, which is usually problematic. There is an increased desire for automated "self-star" configuration, management, and failure-recovery of the data-centre infrastructure, but many traditional techniques scale much worse than linearly as the number of nodes to be managed increases. As the number of nodes in a median-sized data-centre looks set to increase by two or three orders of magnitude in coming decades, it seems reasonable to attempt to explore and understand the scaling properties of the data-centre middleware before such data-centres are constructed. In [1] we presented SPECI, a simulator that predicts aspects of large-scale data-centre middleware performance, concentrating on the influence of status changes such as policy updates or routine node failures. [...]. In [1] we used a first-approximation assumption that such subscriptions are distributed wholly at random across the data centre. In this present paper, we explore the effects of introducing more realistic constraints to the structure of the internal network of subscriptions. We contrast the original results [...] exploring the effects of making the data-centre's subscription network have a regular lattice-like structure, and also semi-random network structures resulting from parameterised network generation functions that create "small-world" and "scale-free" networks. We show that for distributed middleware topologies, the structure and distribution of tasks carried out in the data centre can significantly influence the performance overhead imposed by the middleware.

cs.DC↗