Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Asma’ Khoirunnisa’ has a total of 25 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Is Self-Perceived Data Literacy Psychometrically Defensible? Preliminary Evidence from the Data Literacy Self-Perception Questionnaire

    by Asma’ Khoirunnisa’, Heri Retnawati, Risky Setiawan, Edi Istiyono, Widihastuti, Samsul Hadi, Hanifah Mar’Atush Shalihah

    Self-perception questionnaires offer an efficient way to map students’ data-literacy perceptions, but evidence from a single administration cannot establish broad psychometric validity. This study examined preliminary evidence for the 53-item Data Literacy Self-Perception Questionnaire (DLSPQ), which covers six dimensions: Basic Statistical Concepts, Basic Excel Skills, Excel Functions, Data Analysis and Visualization, Attitudes Toward Data, and Interest in Using Data. Participants were 354 Grade XI students aged 16–17 years from one upper-secondary school in Indonesia. Content evidence was examined through expert judgment and Aiken’s V, followed by item diagnostics, polychoric exploratory factor analysis (EFA), and internal-consistency estimation. Cronbach’s alpha ranged from 0.88 to 0.95 and McDonald’s omega from 0.90 to 0.97 across dimensions. Parallel analysis and the minimum average partial (MAP) criterion converged on a six-factor solution, and the obliquely rotated pattern was broadly consistent with the proposed domains. However, the available analysis record did not include the complete item-level loading matrix, cross-loadings, communalities, factor correlations, or numerical parallel-analysis and MAP outputs; the structural result is therefore treated as provisional rather than confirmatory. Descriptive profiles indicated stronger confidence in routine spreadsheet operations and positive attitudes than in conceptual statistical reasoning, advanced Excel functions, and scatterplot interpretation. Gender profiles are reported descriptively because measurement invariance was not tested. The findings provide preliminary, single-site evidence for group-level diagnostic use and require replication with an independent, more diverse sample and performance-based measures.

  • Open Access

    Data-driven analysis of growth factors in oyster mushroom cultivation: a case study from Indonesia’s market

    by Yosef Budiman, Gilang Adi Prasetyo, Asma’ Khoirunnisa’, Hanifah Mar’ATush Shalihah, Muhamad Riyan Maulana, Yanuar Agung Fadlullah, Sugiri Sugiri

    The oyster mushroom is one of the potential agricultural products that can be developed as an alternative to other agricultural products, to maintain Indonesia's economic condition. However, the production of oyster mushrooms remains low and falls short of the minimum amount of market demand. This study employs a machine learning (ML)–based approach to identify the key parameters influencing oyster mushroom production rates. Recursive feature elimination (RFE) was applied to reduce the initial 19 features to nine, enabling faster processing while maintaining high predictive accuracy. The results showed that agricultural features showed a high contribution rather than environmental, economic, and demographic features. Furthermore, these parameters were related to the train-test analysis to visualize the statistical analysis shown by the best method, adaptive boosting (AdaBoost), with coefficient of determination (R2), mean squared error (MSE), and mean absolute error (MAE) values of 0.997575, 0.009841, and 0.085884, respectively. Related research relevant to the research findings was analyzed to validate that agricultural product features affect the decline of oyster mushroom production. Other supported research conducted by integrating real-time analysis and twin digital models, which can enhance substrate quality.

  • Open Access

    THE EFFECT OF SAMPLE SIZE ON THE STABILITY OF XGBOOST MODEL PERFORMANCE IN PREDICTING STUDENT STUDY PERIOD

    by Muhammad Lintang Damar Sakti, Jailani Jailani, Heri Retnawati, Kana Hidayati, Nur Hadi Waryanto, Zulfa Safina Ibrahim, Asma’ Khoirunnisa, Firdaus Amruzain Satiranandi Wibowo, Miftah Okta Berlian, Angella Ananta Batubara

    Student success can be defined based on the period of study taken until graduation from college. Machine learning can be used to predict the factors that are thought to influence student success. To achieve optimal machine learning model performance, attention is needed on the sample size. This study aims to determine the effect of student sample size on the stability of model performance to predict student success. This research is quantitative. The data used is student data from a university in Yogyakarta from 2014 to 2019, totaling 19061 students. The target variable is the student study period in months, while the predictor variables are college entrance pathways, GPA from semester 1 to semester 6, and family socioeconomic conditions based on the father’s and mother’s income. This research uses the XGBoost model with the best hyperparameters and the bootstrap approach. Bootstrapping was performed on the original data by sampling twenty different sample sizes: 250, 500, 750, 1000, 1250, 1500, 1750, 2000, 2250, 2500, 2750, 3000, 3250, 3500, 3750, 4000, 4250, 4500, 4750, and 5000. The resulting bootstrap samples were replicated ten times. Model performance evaluation uses the Root Mean Square Error (RMSE) value. The result of this research is the XGBoost model with the best hyperparameters, obtained through the training data division scheme of 90% and testing data of 10%, which has the smallest RMSE value of 8.318. The model uses the best hyperparameters: n_estimators of 75, max_depth of 8, min_child_weight of 5, eta of 0.07, gamma of 0.2, subsample of 0.8, and colsample_bylevel of 1. The XGBoost model with optimal hyperparameters demonstrates peak performance stability at a sample size of 1750 students, as evidenced by consistent RMSE values across 10 bootstrap replications, confirming that this data quantity provides the ideal balance between prediction accuracy and stability for estimating study duration.


  • Open Access

    Sentiment Analysis of Consumer Acceptance of Honda’s Digital Marketing Strategy Using Lexicon-Based Algorithm

    by Bartolomius Dias, Asma’ Khoirunnisa’, Yosef Budiman, Setiyawami Setiyawami

    This study analyzes customer sentiment toward Honda’s digital marketing strategy via the Wahana Honda application. A total of 2,000 customer reviews were collected from the Google Play Store using web-scraping techniques. Text data underwent preprocessing (e.g. cleansing, tokenization, stop-word removal, stemming, and translation into English). Sentiment classification using a lexicon-based approach revealed that 56.7% of reviews were positive, 20.8% neutral, and 22.5% negative. The model demonstrated high precision in identifying negative sentiment, though it showed limitations in classifying neutral opinions due to linguistic ambiguity. These findings highlight the need for more adaptive sentiment models and offer strategic insights for Honda’s digital marketing. Specifically, the analysis can help prioritize improvements in app functionality, excellence service priority, enhance personalized customer engagement, and shape targeted digital marketing strategies based on real user feedback. Leveraging these insights enables Honda to optimize user experience, increase retention, and align digital campaigns with customer expectations.


  • Open Access

    MODIFICATION OF POCUNG SONG IN ETHNOMATHEMATICS-BASED LEARNING FOR THE TOPICS OF LENGTH MEASUREMENTS

    by Lilis Risma Putri, Hanifah Mar'ATush Shalihah, Asma' Khoirunnisa', Heru Sukoco

    Learning in measuring units of length is basically inseparable from familiar objects in the society, one of which is traditional songs. This study was aimed at finding whether using the tune of a traditional song in Yogyakarta, namely Pocung song, and modifying the lyrics with mathematical contents can facilitate students to think mathematically. This study used the pre-experimental design method with the one-group pre-test-posttest design model. The research subjects were 10 3rd grade students of a cultural elementary school in Gunungkidul. The results showed that the modification of the Pocung song in ethnomathematics-based mathematics learning in the length-unit competencies was developed in an audio-visual form which could be accessed via the YouTube and the implementation ofthe Pocung song modification in ethnomatematics-based learning in the length-unit competencies had a positive influence on students’ learning achievements.


  • Perbandingan Analisis Agglomerative Hierarchical Clustering Berdasarkan Indikator Pendidikan di Provinsi Jawa Barat

    by Asma' Khoirunnisa', Firdaus Amruzain Satiranandi Wibowo, Kimiantini Kismiantini

    Bidang pendidikan pada saat ini merupakan salah satu prioritas yang menjadi urgensi pemerintah. Berdasarkan Badan Pusat Statistik 2021, tercatat bahwa angka putus sekolah di Provinsi Jawa Barat menduduki peringkat pertama dari 34 provinsi di Indonesia. Fenomena ini tentunya menurunkan indeks pendidikan serta bertolak belakang dengan intensi penting yang menjadi fokus pembangunan pemerintah yakni pemerataan pendidikan. Pada penelitian ini akan dilakukan pemetaan wilayah di Provinsi Jawa Barat berdasarkan indeks pendidikan sehingga didapatkan perencanaan dan kebijakan pemerintah yang lebih akurat. Salah satu algoritma yang dapat digunakan dalam pengelompokan kabupaten berdasarkan indikator pendidikan adalah Agglomerative Hierarchical Clustering (AHC). Beberapa metode pengelompokkan yang digunakan pada penelitian ini adalah AHC dengan metode single linkage, metode complete linkage, metode average linkage, dan metode ward. Hasil dari penelitian ini adalah perbandingan metode yang dilakukan berdasarkan nilai koefisien korelasi cophenetic. Berdasarkan analisis pada penelitian ini, diketahui gambaran pada sektor pendidikan di Provinsi Jawa Barat tahun 2021, yaitu masih terdapat indikasi belum tercapainya pemerataan pendidikan khususnya pada jenjang sekolah menengah atas hingga perguruan tinggi serta terpusatnya jumlah sekolah dan guru di klaster tertentu. Tinjauan indeks pendidikan ini dibentuk menjadi 4 klaster pada setiap masing-masing metode. Berdasarkan hasil perbandingan nilai koefisien korelasi cophenetic dari keempat metode pada analisis AHC, metode ward adalah metode terbaik dengan koefisien korelasi cophenetic terbesar (0,952).