Theses and Dissertations at Montana State University (MSU)
Permanent URI for this collectionhttps://scholarworks.montana.edu/handle/1/733
Browse
13 results
Search Results
Item An evaluation of graph representation of programs for malware detection and categorization using graph-based machine learning methods(Montana State University - Bozeman, College of Engineering, 2023) Pearsall, Reese Andersen; Chairperson, Graduate Committee: Clemente IzurietaWith both new and reused malware being used in cyberattacks everyday, there is a dire need for the ability to detect and categorize malware before damage can be done. Previous research has shown that graph-based machine learning algorithms can learn on graph representations of programs, such as a control flow graph, to better distinguish between malicious and benign programs, and detect malware. With many types of graph representations of programs, there has not been a comparison between these different graphs to see if one performs better than the rest. This thesis provides a comparison between different graph representations of programs for both malware detection and categorization using graph-based machine learning methods. Four different graphs are evaluated: control flow graph generated via disassembly, control flow graph generated via symbolic execution, function call graph, and data dependency graph. This thesis also describes a pipeline for creating a classifier for malware detection and categorization. Graphs are generated using the binary analysis tool angr, and their embeddings are calculated using the Graph2Vec graph embedding algorithm. The embeddings are plotted and clustered using K-means. A classifier is then built by assigning labels to clusters and the points within each cluster. We collected 2500 malicious executables and 2500 benign executables, and each of the four graph types is generated for each executable. Each is plugged into their own individual pipeline. A classifier for each of the four graph types is built, and classification metrics (e.g. F1 score) are calculated. The results show that control flow graphs generated from symbolic execution had the highest F1 score of the four different graph representations. Using the control flow graph generated from symbolic execution pipeline, the classifier was able to most accurately categorize trojan malware.Item Variety of strategies used to teach data analysis and conclusion writing in Freshmen Physics(Montana State University - Bozeman, College of Letters & Science, 2019) Shaaban, Lori; Chairperson, Graduate Committee: Greg FrancisDue to our data driven society, students should understand how to make sense of graphs and be able to apply them. Educators need to teach students how to analyze data, communicate that understanding, and pose new questions. The Next Generation Science Standards (NGSS) places a heavy importance on analyzing and interpreting data, constructing explanations, and engaging in an argument from evidence due to an increasing need for this skill in the labor force. Two classes of freshmen physics learned techniques in reading, analyzing, and interpreting data to understand physics concepts. They were taught how to spot trends in data tables as well as graphs and used www.desmos.com to find lines of best fit equations. They tried to understand what the equation represented and why the phenomena occurred. Then using their equation, they made a prediction and explained the reasons for their confidence in that prediction. Lastly, they thought of new experiments they could do based on this latest information and how businesses could use data similar. This is a modified version of the Claim-Evidence-Reasoning (CER) conclusion in science classrooms. Since many of the lab reports were done as a group, the action research assessed in this document was not done from students' experiments, but data provided on students' individual tests. Pre- and post-tests, surveys, interviews, and group discussions were reviewed. It was found many students began the course with the ability to make a prediction based on an equation. They quickly figured out how to spot patterns in the data to make a claim. However, the most challenging for students was to explain the phenomena and provide confidence in their prediction. Students did not say any one component was much harder or important than another. Students were overconfident in their ability to explain their confidence scientifically throughout the year. In all, students found a variety of activities helpful as they continued to grow throughout the year.Item Energy and work instructional strategies in general physics(Montana State University - Bozeman, College of Letters & Science, 2018) Tschanz, Chad M.; Chairperson, Graduate Committee: Greg FrancisIntroductory physics students have demonstrated unsatisfactory learning gains after traditional energy instruction. Energy instruction was adapted to include pictorial system diagrams, energy bar graphs, and interactive physical modeling. Learning gains of students who received adapted instruction was compared to students who received traditional instruction. Students who participated in the adapted instruction showed slightly greater gains in interpretation of lab data compared to those students who received traditional instruction. There was no significant difference in gains noticed on traditional assessments of work and energy concepts. No significant differences in preference for student-led interactive physical modeling over traditional instruction was found.Item Graphing and estimation as tools to improve critical thinking in high school chemistry students(Montana State University - Bozeman, College of Letters & Science, 2018) Pelliccia, Chris A.; Chairperson, Graduate Committee: Greg FrancisHigh school conceptual chemistry students engaged in daily estimation activities and frequent in-class graphing to practice mathematical reasoning, argumentation, and visual analysis. Students applied those skills by regularly creating and analyzing graphs using both real-world and lab-generated data sets. Self-confidence surveys, performance assessments, and presentations were used as data collection instruments. Data were processed using quantitative and qualitative analysis strategies. The results suggested that students improved their abilities to create and interpret graphs using mathematical reasoning and visual analysis, key components of critical thinking. Interview data suggests no overall change in student attitude towards the utility of graphs as a means of conveying information.Item Bellwork as a strategy to increase student's ability to analyze graph and chart data(Montana State University - Bozeman, College of Letters & Science, 2018) Doup, Jamie R.; Chairperson, Graduate Committee: Greg FrancisStudents leaving high school should possess certain tasks to be successful in the world. Reading, writing, and basic math skills seem obvious. However, in our ever-changing world where science and technology are advancing at a rapid pace, any adult consumer needs to be able to read data displayed as a chart or graph. This action research-based classroom project utilized the ACT science test as the basis for graph literacy. Data from student pre-and posttest scores are compared after three weeks of bellwork, lasting only 5-7 minutes per day. The small but significant increases in student confidence and skills, increased their number of completed questions and overall scores.Item The impacts of learning with multiple representations in a high school physics classroom(Montana State University - Bozeman, College of Letters & Science, 2018) Carroll, Joshua George; Chairperson, Graduate Committee: Greg FrancisStudent learning in physics takes on many forms. Equations, diagrams, graphs and words all can be used to describe physical phenomena. Constructing descriptions of physical situations with these representations and focusing on their correct usage is a goal of all physics instruction. Teaching students the strengths of these representations in communicating conceptual ideas and guiding students in usefulness of representations in problem-solving will is how this goal can be accomplished. This study investigated whether learning physics with an emphasis on multiple representations had an effect on student conceptual understanding, student problem solving, and student attitude in a high school introductory physics classroom. Through instruction that emphasized student practice with the representations themselves, developing a problem-solving method that included multiple representations, and assessment of quality of representations, students learned a unit on unbalanced forces. Results and conclusions showed low increases in student conceptual understanding and problem-solving ability. Student attitude improved in regard to their view of its application in the real-world. This classroom research project showed that students should be well-versed in all representations in order to achieve high gains in learning and attitude.Item To what extent do graphic organizers influence the academic achievement of ninth-grade chemistry students?(Montana State University - Bozeman, College of Letters & Science, 2017) Kummari, Veeraiah; Chairperson, Graduate Committee: Greg FrancisThis study investigated the effect of graphic organizers on the academic achievement of ninth-grade chemistry students (N = 22). Data was collected on their ability to design and use concept maps, mind maps, Vee diagrams, and Venn diagrams. The results showed that there was a positive correlation between the use of graphic organizers and students' test scores. The findings also revealed that the students were more engaged and took greater responsibility for their learning in the post-mapping period.Item Improving visual data literacy skills of high school earth and space science students by weekly data analysis curriculum(Montana State University - Bozeman, College of Letters & Science, 2017) Suzak, Miranda G.; Chairperson, Graduate Committee: Greg FrancisStudents must be able to interpret and analyze visual data to be successful in science classes, on high stakes assessment testing, and in their adult life. Visual data includes graphs, tables, charts, and diagrams. Traditionally, students receive instruction and practice using visual data literacy skills repeatedly in elementary school, middle school, and high school. This is a challenging skill set, as it requires scientific logical thinking and interpretation of abstract information. Students practice and grow as they continue through school, but continually seem to struggle to achieve mastery of visual data literacy. This project investigated how to improve student understanding, analysis, and use of visual data by focusing on a weekly treatment of Interpreting the Graph. Each week students focused on a piece of visual data that related to current class curriculum. Students worked in lab groups to understand and analyze the data. Class discussion allowed students to share their thoughts and discuss areas that were hard to understand. Data collected for this project included a student survey before the treatment process that assessed student perception of their own abilities and skill in visual data literacy. Students then took a pretest with data literacy questions using graphs, tables, charts, and diagrams. Each week for six weeks students filled out an Interpreting the Graph student worksheet. Each week students also recorded their weekly progress and overall class progress in a graphing packet. At the end of the six week treatment process students took a posttest. Students filled out a student survey about their growth and confidence in data literacy skills through the project, and a selection of students were interviewed about their growth in visual data literacy skills. The intervention seemed to show a positive impact on student learning and abilities. Students showed a 16.7% gain between pretest and posttest assessments. Through the course of the six week intervention, students showed an increase in confidence in visual data literacy skills, as well as showing improvement on the posttest assessment. Students improved their abilities to identify independent and dependent variables of an experiment. However, students showed limited growth in their ability to explain theoretical experimental methods used to create visual data seen in class, and in their ability to write explanations for patterns in visual data using numerical evidence from the data. The key findings showed that students seemed to benefit from learning a methodical analysis procedure to work through interpreting visual data. Students also benefited from participating in regular class discussion about visual data as it pertains to the curriculum. These findings were used in continued discussion with students to help them focus on their own learning. The methods from the intervention were used in curriculum planning to change how visual data analysis was taught in Earth and Space Science classes in future years and shared with other secondary science classes to increase student skill in visual data interpretation and analysis within our school district.Item The relationship and importance of graphical representation to learning(Montana State University - Bozeman, College of Education, Health & Human Development, 1965) Parsons, Henry LehnerItem MAXPLANAR : a graphical software package for testing maximal planar subgraph algorithms(Montana State University - Bozeman, College of Engineering, 1996) Zhao, Kedan