Chinese Comma Disambiguation in Math Word Problems Using SMOTE and Random Forests

Natural language understanding technologies play an essential role in automatically solving math word problems. In the process of machine understanding Chinese math word problems, comma disambiguation, which is associated with a class imbalance binary learning problem, is addressed as a valuable instrument to transform the problem statement of math word problems into structured representation. Aiming to resolve this problem, we employed the synthetic minority oversampling technique (SMOTE) and random forests to comma classification after their hyperparameters were jointly optimized. We propose a strict measure to evaluate the performance of deployed comma classification models on comma disambiguation in math word problems. To verify the effectiveness of random forest classifiers with SMOTE on comma disambiguation, we conducted two-stage experiments on two datasets with a collection of evaluation measures. Experimental results showed that random forest classifiers were significantly superior to baseline methods in Chinese comma disambiguation. The SMOTE algorithm with optimized hyperparameter settings based on the categorical distribution of different datasets is preferable, instead of with its default values. For practitioners, we suggest that hyperparameters of a classification models be optimized again after parameter settings of SMOTE have been changed.

The Contemporary Significance of the Early Accounting Teaching Methods

In the light of global trends, economic operators cannot withstand change. At first, computer-driven technologies replaced only routine tasks, which were easily programmed using algorithms. However, as a result of technological development, artificial intelligence, machine learning, the internet and big data, machines have acquired an understanding of non-routine tasks. They have become autonomous, and are now capable of solving more and more complex tasks.The job opportunities offered by the new digitalized world ask for new competencies developed by the education system. Our research examines 20 th-century teaching methods based on final exams made public, then compares and contrasts them to 21 st-century teaching materials and examination methods. One of the significantobservations is that between 1900 and 1918, the final exam in mathematics contained only word problems. The majority of the 223 problems available, 57 by number, focused on “capital, interest, benefit, loan, sales and purchase.” The wording of the problemsreflects the (actual) problems and events of the time. Although case studies are still present, most problems are the “calculate, solve, circle, underline” type. Example problems are often provided; thus, problem-solving turns into a routine task. The earlier method yet based on word problems inspired by our everyday economic reality might prove helpful in developing problem-solving skills, in reducing reading comprehension difficulties present at all levels of education as well as in indirectly raising awareness of today’s environmental, personal finance, issues.

THE EFFECT OF READING COMPREHENSION SKILLS ON ABILITY TO SOLVE MATH WORD PROBLEMS TO FOURTH GRADE STUDENTS OF PUBLIC ELEMENTARY SCHOOLS IN KEBUMEN SUB-DISTRICT IN ACADEMIC YEAR OF 2020/2021

<p>The research aimed to determine the effect of reading comprehension skills on the ability to solve math word problems and to determine the contribution of reading comprehension skills on the ability to solve math word problems to fourth grade students of public elementary schools in Kebumen Sub-district in academic year of 2020/2021. The research was a quantitative research with regression method. The population were all fourth grade students in Kebumen Sub-district. The samples were 324 fourth grade. Sampling method was probability sampling technique. The data collection technique used tests. The results showed that there was a positive and significant effect between reading comprehension skills on the ability to solve math word problems to fourth grade students of public elementary schools in Kebumen Sub-district in academic year of 2020/2021. The value of determination coefficient (R Square) was 0.214. The reading comprehension skills contributed 21.3% on the ability to solve math word problems and the remaining 78.7% was influenced by other factors. </p><p> </p><p> </p><p>Penelitian ini bertujuan untuk mengetahui adanya pengaruh kemampuan membaca pemahaman terhadap kemampuan menyelesaikan soal cerita matematika dan mengetahui besarnya sumbangan kemampuan membaca pemahaman terhadap kemampuan menyelesaikan soal cerita matematika siswa kelas IV SD Negeri se-Kecamatan Kebumen Tahun Ajaran 2020/2021. Penelitian ini merupakan penelitian kuantitatif dengan metode regresi. Populasi penelitian ini adalah seluruh siswa kelas IV SDN se-Kecamatan Kebumen. Sampel pada penelitian ini berjumlah 324 siswa yang dipilih secara acak dengan teknik probability sampling. Teknik pengumpulan data yang digunakan dalam penelitian ini berupa tes. Hasil penelitian ini menunjukkan terdapat pengaruh positif dan signifikan antara kemampuan membaca pemahaman terhadap kemampuan menyelesaikan soal cerita matematika siswa kelas IV SDN se-Kecamatan Kebumen tahun ajaran 2020/2021 dengan nilai (R Square) sebesar 0,214. Kemampuan membaca pemahaman memberikan sumbangan terhadap kemampuan menyelesaikan soal cerita matematika sebesar 21,3 % dan sisanya sebesar 78,7 % dipengaruhi oleh faktor lain.</p><div> </div>

Tahapan Pemecahan Masalah dengan Diagram Bar pada Soal Cerita

Word problems are still one of the causes of students' difficulties in learning mathematics. The four stages of problem solving suggested by Polya are: (1) understanding the problem, (2) planning a strategy, (3) implementing the strategy, and (4) checking and interpreting the results. One of the strategies to solve word problems that can be applied is by making diagram. The bar diagram model is a form of visual representation that can help students to solve word problems. The use of bar diagram in solving word problems can be a strategy for students to bridge the transition from understanding problems and making mathematical expressions. The process of creating a bar diagram model represents a visualization of known and unknown information and helps to identify how this information can be used to solve the problems. This paper discusses problem solving stages using bar diagram to solve word problems.

Teaching Students With Mild Intellectual Disability to Solve Word Problems Using Schema-Based Instruction

This study, which used a multiple baseline across students’ design, examines the effectiveness of a modified schema-based instructional approach to improve the mathematical word problem-solving performance of three students with mild intellectual disability, two of them with autism spectrum disorder. Following the intervention, the three students improved their performance when solving addition and subtraction change word problems; however, their performance was inconsistent with change word problems. The effects of the instruction were generalized to two-step addition and subtraction word problems for the three participants. Moreover, the results were generalized to an untrained setting and were maintained 8 weeks after the instruction. The implications of these findings for teaching problem-solving skills to students with intellectual disability are discussed.

When does the story matter? No evidence for the foregrounding hypothesis in math story problems

Math story problems are difficult for many solvers because comprehension of mathematical and linguistic content must occur simultaneously. Across two studies, we attempted to conceptually replicate and extend findings reported by Mattarella-Micke and Beilock (2010, https://doi.org/10.3758/PBR.17.1.106) and Jarosz and Jaeger (2019, https://doi.org/10.1002/acp.3471). Mattarella-Micke and Beilock found that multiplication word problems in which an irrelevant number was associated with the protagonist of the problem (i.e., foregrounded in the text) were solved less accurately than problems in other conditions. Jarosz and Jaeger used similar materials but tested the more general inconsistent-operations hypothesis that association with the protagonist would interfere with multiplication whereas dissociation would interfere with division. They found partial support: When division problems were primed with dissociative scenarios, solvers made more errors, but they failed to replicate the associative findings for multiplication. In the present research, we conducted two studies (Ns = 205 and 359), in which we similarly manipulated whether irrelevant content was associated with or dissociated from the story protagonist. In these studies, we did not find support for either the foregrounding or inconsistent-operations hypotheses. Exploratory error analyses suggested that solvers’ errors were most often the result of calculation difficulties or inappropriate operation choices and were unrelated to the presence of associative or dissociative story elements. Our careful implementation of this manipulation and much greater power to detect effects suggests that the association manipulation in irrelevant text does not influence adults’ performance on simple math story problems.