This project is for CIVE 700

• C30 RC and MW classified as severe. However, from images seems to be light.

• The document starts by discussing the importance of rapid seismic vulnerability assessment to understand a community's resilience to earthquakes, focusing on the aftermath of the Haiti 2010 earthquake.

• The data from the earthquake reconnaissance in Haiti is analyzed to understand the damage to reinforced concrete structures and roofing materials, with a focus on light and severe damage.

• Descriptive statistics are provided for different damage categories, and a breakdown of the data on RC damage is given based on inspection teams' reports.

• Data pre-processing steps are outlined, including the exclusion of certain classes of data due to low representation. Machine learning methods like logistic regression and support vector machines (SVMs) are explained in the context of earthquake damage assessment.

• Initial results from logistic regression and SVM models are presented, showing varying levels of accuracy in classifying damage classes using different methods.

• The document discusses the limitations of the current models and proposes future work to improve model performance, including feature selection, outlier removal, and iterative refinement of support vectors.

• A list of references cited in the document is provided, including research papers and resources related to seismic vulnerability assessment.