The work is related to the creation of an effective algorithm that allows you to identify stock transactions with a sign of manipulation. The fundamental reasons for the research are the growing volume of the securities market and the emergence of digital technologies. The chapter presents a general statement of the problem. The criterion used is based on the concept of information entropy. It is calculated as a function of the various time series patterns identified during the trading process. A sign of manipulation is the probability that a sequence of length m will be part of a longer pattern. To compare patterns, the finite deterministic automata method procedure was used. The emergence of patterns in stock trading is usually associated with attempts to manipulate prices in order to change the latter in a favorable direction. The proposed algorithm is based solely on real trading transactions and, of course, does not cover all possible types of manipulation. However, it does not depend on the type of traded instrument. The chapter presents the results of calculations demonstrating the effectiveness of the algorithm.