In order to efficiently analyse the vast amount of data generated by solar space missions and ground-base instruments, modern machine learning techniques such as decision trees, support vector machines (SVMs) and neural networks can be very useful. In this paper we present initial results from using a convolutional neural network (CNN) to analyse observations from the Atmospheric Imaging Assembly (AIA) in the 1600A wavelength. The data is pre-processed to locate flaring regions where flare ribbons are visible in the observations. The CNN is created and trained to automatically analyse the shape and position of the flare ribbons, by identifying whether each image belongs into one of four classes: two-ribbon flare, compact/circular ribbon flare, limb flare or quiet Sun, with the final class acting as a control for any data included in the training or test sets where flaring regions are not present. The network created can classify flare ribbon observations into any of the four classes with a final accuracy of 94%. Initial results show that most of the images are correctly classified with the compact flare class being the only class where accuracy drops below 90% ad some observations are wrongly classified as belonging to the limb class.

We present analysis of MACHO Alert 95-30, a dramatic gravitational microlensing event towards the Galactic bulge whose peak magnification departs significantly from the standard point-source microlensing model. Alert 95-30 was observed in real-time by the Global Microlensing Alert Network (GMAN), which obtained densely sampled photometric and spectroscopic data throughout the event. We interpret the light-curve ``fine structure'' as indicating transit of the lens across the extended face of the source star. This signifies resolution of a star several kpc distant. We find a lens angular impact parameter theta_{min}/theta_{source} = 0.715 +/- 0.003. This information, along with the radius and distance of the source, provides an additional constraint on the lensing system. Spectroscopic and photometric data indicate the source is an M4 III star of radius 61 +/- 12 Rsun, located on the far side of the bulge at 9 kpc. We derive a lens angular velocity, relative to the source, of 21.5 +/- 4.9 km/s/kpc, where the error is dominated by uncertainty in the source radius. Likelihood analysis yields a median lens mass of 0.67{+2.53}{-0.46} Msun, located with 80% probability in the Galactic bulge at a distance of 6.93{+1.56}{-2.25} kpc. If the lens is a main-sequence star, we can include constraints on the lens luminosity. This modifies our estimates to M_lens = 0.53{+0.52}{-0.35} Msun and D_lens = 6.57{+0.99}{-2.25} kpc. Spectra taken during the event show that the absorption line equivalent widths of H alpha and the TiO bands near 6700 A vary, as predicted for microlensing of an extended source. This is most likely due to center-to-limb variation in the stellar spectral lines. These data demonstrate the feasibility of using microlensing limb crossings as a tool to probe stellar atmospheres directly.