As we rely more on satellites, communication systems and space research, the importance of space weather is increasing continuously. There are many space missions and ground based observatories providing continuous observation of the Sun at many different wavelengths to supply the demand for space weather forecast and research. All the forecasting strategies highly depend on experience of solar physicists and done manually. The results differ from observatories to observatories and subjective. There is a need for automated analysis of Sun and space weather forecasting. The solar activity is the driver of space weather. Thus it is important to be able to predict the violent eruptions such as coronal mass ejections and solar flares. In this book a hybrid system combining image processing and machine learning techniques for the automated short-term prediction of solar flares is presented. The system can also detect, group, and classify sunspots using solar images. The algorithms, implementation, and results are explained in this work.

Space weather has become an international issue due to the catastrophic impactit can have on modern societies. Solar flares are one of the major solar activities thatdrive space weather and yet their occurrence is not fully understood. Research isrequired to yield a better understanding of flare occurrence and enable the developmentof an accurate flare prediction system, which can warn industries most at risk to takepreventative measures to mitigate or avoid the effects of space weather. This thesisintroduces novel technologies developed by combining advances in statistical physics, image processing, machine learning, and feature selection algorithms, with advances insolar physics in order to extract valuable knowledge from historical solar data, related toactive regions and flares. The aim of this thesis is to achieve the followings: i) Thedesign of a new measurement, inspired by the physical Ising model, to estimate themagnetic complexity in active regions using solar images and an investigation of thismeasurement in relation to flare occurrence. The proposed name of the measurement isthe Ising Magnetic Complexity (IMC). ii) Determination of the flare predictioncapability of active region properties generated by the new active region detectionsystem SMART (Solar Monitor Active Region Tracking) to enable the design of a newflare prediction system. iii) Determination of the active region properties that are mostrelated to flare occurrence in order to enhance understanding of the underlying physicsbehind flare occurrence. The achieved results can be summarised as follows: i) The newactive region measurement (IMC) appears to be related to flare occurrence and it has apotential use in predicting flare occurrence and location. ii) Combining machinelearning with SMART's active region properties has the potential to provide moreaccurate flare predictions than the current flare prediction systems i.e. ASAP(Automated Solar Activity Prediction). iii) Reduced set of 6 active region propertiesseems to be the most significant properties related to flare occurrence and they canachieve similar degree of flare prediction accuracy as the full 21 SMART active regionproperties. The developed technologies and the findings achieved in this thesis willwork as a corner stone to enhance the accuracy of flare prediction; develop efficientflare prediction systems; and enhance our understanding of flare occurrence. Thealgorithms, implementation, results, and future work are explained in this thesis.

Coronal Mass Ejections (CMEs) and solar flares are energetic events taking place at the Sun that can affect the space weather or the near-Earth environment by the release of vast quantities of electromagnetic radiation and charged particles. Solar active regions are the areas where most flares and CMEs originate. Studying the associations among sunspot groups, flares, filaments, and CMEs is helpful in understanding the possible cause and effect relationships between these events and features. Forecasting space weather in a timely manner is important for protecting technological systems and human life on earth and in space. The research presented in this thesis introduces novel, fully computerised, machine learning-based decision rules and models that can be used within a system design for automated space weather forecasting. The system design in this work consists of three stages: (1) designing computer tools to find the associations among sunspot groups, flares, filaments, and CMEs (2) applying machine learning algorithms to the associations' datasets and (3) studying the evolution patterns of sunspot groups using time-series methods. Machine learning algorithms are used to provide computerised learning rules and models that enable the system to provide automated prediction of CMEs, flares, and evolution patterns of sunspot groups. These numerical rules are extracted from the characteristics, associations, and time-series analysis of the available historical solar data. The training of machine learning algorithms is based on data sets created by investigating the associations among sunspots, filaments, flares, and CMEs. Evolution patterns of sunspot areas and McIntosh classifications are analysed using a statistical machine learning method, namely the Hidden Markov Model (HMM).

This study looked at observational and theoretical studies of flare physics, at quests for flare precursors, and at mathematical models for combining masses of predictive information. We also looked at the worldwide effort to gather and share timely data and combine it with knowledge and experience to forecast solar flares and their effects. Topics include: Long-lived, large-scale magnetic and velocity fields; Magnetic-energy buildup in an active region; Flare initiation; Flare precursors -- Filament activation, Preflare brightening, Magnetic shear, and Emerging and cancelling magnetic flux; Quantitative prediction; Operational solar flare prediction; Forecast evaluation.