Tunnel squeezing is a time dependent process that typically occurs in weak or over-stressed rock masses, 2 significantly influencing the budget and time of tunnel construction. This paper presents a new framework to 3 probabilistically predict the potential squeezing intensity, and to dynamically update the prediction during 4 construction based on the sequentially revealed ground information. An extensively well documented database, 5 which contains quantitative data from 154 squeezing sections with 95 unpublished inventories is established. A 6 Decision Tree method is employed to train a probabilistic multi-classification model to predict the tunnel 7 squeezing intensity. The trained classifier is then integrated with a Markovian geologic model, which features 8 embedded Bayesian updating procedures, to achieve a dynamic prediction on the state probabilities of the 9 geologic parameter within the Markovian geologic model and the resulting squeezing intensity during 10 excavation. An under-construction tunnel case -Miyaluo #3 tunnel-is used to illustrate the proposed 11 framework. Results show that the Decision Tree classifier, as opposed to other black-box models, is easily to be 12 interpreted. It provides reliable predictive accuracy while leading to insights into the understanding of squeezing 13 problem. The strength-stress ratio (SSR) is suggested to be the most important factor. Moreover, the 14 implementation of the updating procedures is efficient since only simple field test (eg. Point Load index or 15 Schmidt rebound index) is required. Multiple rounds of predictions within the updating process allow different 16 levels of prediction, for example long-range, short-term, or immediate, to be extracted as useful information 17 towards the decision-making of construction operations. Therefore, this framework can serve as a pragmatic tool 18 to assist the selection of optimal primary-support and other construction strategies based on the potential 19 squeezing risk.