You have to be careful here because you can potentially be talking about two different phenomena. As @vchris_ngs pointed out in his answer, there is an immunological phenomenon described by this term. But given your question and the previous question you had about Hypermutation, I believe you are probably asking about hypermutation in the context of cancer.
Some review papers that may be of interest can be found here and here. Briefly all mutations that occur in non-germline cells during the lifetime of an individual are somatic mutations (as they occur in somatic tissue). In the context of cancer we are generally talking specifically about mutations in the tumour as opposed to the individual's germline. Hypermutation isn't an observed phenomenon in some tumours, it is basically the cancer version of a mutator phenotype and is most often associated with some defect, in the tumour, in DNA repair processes such that the tumour has a higher mutation rate compared to other tumours of the same type. It is usually a pretty significant difference in mutation rate, such that these tumours have far more somatic mutations than what is expected.
You can easily find the definition of it in wiki. Somatic mutations are more classical mutations that are acquired during the evolution of a disease and can be one of the drivers while in case of somatic hyper mutation if they are mistargeted they lead to B-Cell lymphomas since they are is a cellular mechanism by which the immune system adapts to the new foreign elements that confront it (e.g. microbes), as seen during class switching (wiki source). It is more like a phenomenon happening while somatic mutation is a term associated to classify anything that is not in the germline and has been acquired in course of evolution that might have dramatic effect on the phenotype or typically can drive the disease.