Service complaint identification in hotel social media: A two-step classification approach

Author(s):  
Jiahua Jin ◽  
Lu Lu

Hotel social media provides access to dissatisfied customers and their experiences with services. However, due to massive topics and posts in social media, and the sparse distribution of complaint-related posts and, manually identifying complaints is inefficient and time-consuming. In this study, we propose a supervised learning method including training samples enlargement and classifier construction. We first identified reliable complaint and noncomplaint samples from the unlabeled dataset by using small labeled samples as training samples. Combining the labeled samples and enlarged samples, classification algorithms support vector machine and k-nearest neighbor were then adopted to build binary classifiers during the classifier construction process. Experimental results indicate the proposed method can identify complaints from social media efficiently, especially when the amount of labeled training samples is small. This study provides an efficient approach for hotel companies to distinguish a certain kind of consumer complaint information from large number of unrelated information in hotel social media.

Author(s):  
Maria Morgan ◽  
Carla Blank ◽  
Raed Seetan

<p>This paper investigates the capability of six existing classification algorithms (Artificial Neural Network, Naïve Bayes, k-Nearest Neighbor, Support Vector Machine, Decision Tree and Random Forest) in classifying and predicting diseases in soybean and mushroom datasets using datasets with numerical or categorical attributes. While many similar studies have been conducted on datasets of images to predict plant diseases, the main objective of this study is to suggest classification methods that can be used for disease classification and prediction in datasets that contain raw measurements instead of images. A fungus and a plant dataset, which had many differences, were chosen so that the findings in this paper could be applied to future research for disease prediction and classification in a variety of datasets which contain raw measurements. A key difference between the two datasets, other than one being a fungus and one being a plant, is that the mushroom dataset is balanced and only contained two classes while the soybean dataset is imbalanced and contained eighteen classes. All six algorithms performed well on the mushroom dataset, while the Artificial Neural Network and k-Nearest Neighbor algorithms performed best on the soybean dataset. The findings of this paper can be applied to future research on disease classification and prediction in a variety of dataset types such as fungi, plants, humans, and animals.</p>


2016 ◽  
Vol 1 (1) ◽  
pp. 13 ◽  
Author(s):  
Debby Erce Sondakh

Penelitian ini bertujuan untuk mengukur dan membandingkan kinerja lima algoritma klasifikasi teks berbasis pembelajaran mesin, yaitu decision rules, decision tree, k-nearest neighbor (k-NN), naïve Bayes, dan Support Vector Machine (SVM), menggunakan dokumen teks multi-class. Perbandingan dilakukan pada efektifiatas algoritma, yaitu kemampuan untuk mengklasifikasi dokumen pada kategori yang tepat, menggunakan metode holdout atau percentage split. Ukuran efektifitas yang digunakan adalah precision, recall, F-measure, dan akurasi. Hasil eksperimen menunjukkan bahwa untuk algoritma naïve Bayes, semakin besar persentase dokumen pelatihan semakin tinggi akurasi model yang dihasilkan. Akurasi tertinggi naïve Bayes pada persentase 90/10, SVM pada 80/20, dan decision tree pada 70/30. Hasil eksperimen juga menunjukkan, algoritma naïve Bayes memiliki nilai efektifitas tertinggi di antara lima algoritma yang diuji, dan waktu membangun model klasiifikasi yang tercepat, yaitu 0.02 detik. Algoritma decision tree dapat mengklasifikasi dokumen teks dengan nilai akurasi yang lebih tinggi dibanding SVM, namun waktu membangun modelnya lebih lambat. Dalam hal waktu membangun model, k-NN adalah yang tercepat namun nilai akurasinya kurang.


Author(s):  
Seyma Kiziltas Koc ◽  
Mustafa Yeniad

Technologies which are used in the healthcare industry are changing rapidly because the technology is evolving to improve people's lifestyles constantly. For instance, different technological devices are used for the diagnosis and treatment of diseases. It has been revealed that diagnosis of disease can be made by computer systems with developing technology.Machine learning algorithms are frequently used tools because of their high performance in the field of health as well as many field. The aim of this study is to investigate different machine learning classification algorithms that can be used in the diagnosis of diabetes and to make comparative analyzes according to the metrics in the literature. In the study, seven classification algorithms were used in the literature. These algorithms are Logistic Regression, K-Nearest Neighbor, Multilayer Perceptron, Random Forest, Decision Trees, Support Vector Machine and Naive Bayes. Firstly, classification performance of algorithms are compared. These comparisons are based on accuracy, sensitivity, precision, and F1-score. The results obtained showed that support vector machine algorithm had the highest accuracy with 78.65%.


Text Classification plays a vital role in the world of data mining and same is true for the classification algorithms in text categorization. There are many techniques for text classification but this paper mainly focuses on these approaches Support vector machine (SVM), Naïve Bayes (NB), k-nearest neighbor (k-NN). This paper reveals results of the classifiers on mini-newsgroups data which consists of the classifies on mini-newsgroups data which consists a lot of documents and step by step tasks like a listing of files, preprocessing, the creation of terms(a specific subset of terms), using classifiers on specific subset of datasets. Finally, after the results and experiments over the dataset, it is concluded that SVM achieves good classification output corresponding to accuracy, precision, F-measure and recall but execution time is good for the k-NN approach.


Nowadays, internet and social media are play and important role for the business and marketing. Especially, the social media marketing drives the businesses with fierce competition. if there is communication between a large number of customers, it is necessary to have the staff to coordinate thoroughly Resulting in higher expenses as well. Chatbot can be solve this problem by action like a human to deliver a suitable message for their customers. This paper proposes the techniques for analyzing the sentiments that coexist with chat messages or the conversations. Naïve Bayes, K-Nearest Neighbor, and Support Vector Machine techniques were used to classify the sentiments based on Cross-Industry Standard Process for Data Mining. As a result, the highest accuracy is produced by Support Vector Machine with value at 94.60% for improving the chatbot able to communicate effectively with sticker messages.


The world today has made giant leaps in the field of Medicine. There is tremendous amount of researches being carried out in this field leading to new discoveries that is making a heavy impact on the mankind. Data being generated in this field is increasing enormously. A need has arisen to analyze these data in order to find out the meaningful and relevant hidden patterns. These patterns can be used for clinical diagnosis. Data mining is an efficient approach in discovering these patterns. Among the many data mining techniques that exists, this paper aims at analyzing the medical data using various Classification techniques. The classification techniques used in this study include k-Nearest neighbor (kNN), Decision Tree, Naive Bayes which are hard computing algorithms, whereas the soft computing algorithms used in this study include Support Vector Machine (SVM), Artificial Neural Networks (ANN) and Fuzzy k-Means clustering. We have applied these algorithms to three kinds of datasets that are Breast Cancer Wisconsin, Haberman Data and Contraceptive Method Choice dataset. Our results show that soft computing based classification algorithms better classifications than the traditional classification algorithms in terms of various classification performance measures


2019 ◽  
Vol 37 (15_suppl) ◽  
pp. 2029-2029 ◽  
Author(s):  
Estela Pineda ◽  
Anna Esteve-Codina ◽  
Maria Martinez-Garcia ◽  
Francesc Alameda ◽  
Cristina Carrato ◽  
...  

2029 Background: Glioblastoma (GBM) gene expression subtypes have been described in last years, data in homogeneously treated patients is lacking. Methods: Clinical, molecular and immunohistochemistry (IHC) analysis from patients with newly diagnosed GBM homogeneously treated with standard radiochemotherapy were studied. Samples were classified based on the expression profiles into three different subtypes (classical, mesenchymal, proneural) using Support Vector Machine (SVM), the K-nearest neighbor (K-NN) and the single sample Gene Set Enrichment Analysis (ssGSEA) classification algorithms provided by GlioVis web application. Results: GLIOCAT Project recruited 432 patients from 6 catalan institutions, all of whom received standard first-line treatment (2004 -2015). Best paraffin tissue samples were selected for RNAseq and reliable data were obtained from 124. 82 cases (66%) were classified into the same subtype by all three classification algorithms. SVM and ssGEA algorithms obtain more similar results (87%). No differences in clinical variables were found between the 3 GBM subtypes. Proneural subtype was enriched with IDH1 mutated and G-CIMP positive tumors. Mesenchymal subtype (SVM) was enriched in unmethylated MGMT tumors (p = 0.008), and classical (SVM) in methylated MGMT tumors (p = 0.008). Long survivors ( > 30 months) were rarely classified as mesenchymal (0-7.5%) and were more frequently classified as Proneural (23.1-26.). Clinical (age, resection, KPS) and molecular ( IDH1, MGMT) known prognostic factors were confirmed in this serie. Overall, no differences in prognosis were observed between 3 subtypes, but a trend to worse survival in mesenchymal was observed in K-NN (9.6 vs 15 ). Mesenchymal subtype presented less expression of Olig2 (p < 0.001) and SOX2 (p = 0.003) by IHC, but more YLK-40 expression (p = 0.023, SVM). On the other hand, classical subtype expressed more Nestin (p = 0.004) compared to the other subtypes (K-NN). Conclusions: In our study we have not found correlation between glioblastoma expression subtype and outcome. This large serie provides reproducible data regarding clinical-molecular-immunohistochemistry features of glioblastoma genetic subtypes.


Author(s):  
Samiul Azam ◽  
Marina L. Gavrilova

Online social media (OSN) has witnessed a significant growth over past decade. Millions of people now share their thoughts, emotions, preferences, opinions and aesthetic information in the form of images, videos, music, texts, blogs and emoticons. Recently, due to existence of person specific traits in media data, researchers started to investigate such traits with the goal of biometric pattern analysis and recognition. Until now, gender recognition from image aesthetics has not been explored in the biometric community. In this paper, the authors present an authentic model for gender recognition, based on the discriminating visual features found in user favorite images. They validate the model on a publicly shared database consisting of 24,000 images provided by 120 Flickr (image based OSN) users. The authors propose the method based on the mixture of experts model to estimate the discriminating hyperplane from 56 dimensional aesthetic feature space. The experts are based on k-nearest neighbor, support vector machine and decision tree methods. To improve the model accuracy, they apply a systematic feature selection using statistical two sampled t-test. Moreover, the authors provide statistical feature analysis with graph visualization to show discriminating behavior between male and female for each feature. The proposed method achieves 77% accuracy in predicting gender, which is 5% better than recently reported results.


2020 ◽  
Vol 16 (3) ◽  
pp. 155014772091189 ◽  
Author(s):  
Zhen-Wu Wang ◽  
Si-Kai Wang ◽  
Ben-Ting Wan ◽  
William Wei Song

The multi-label classification problem occurs in many real-world tasks where an object is naturally associated with multiple labels, that is, concepts. The integration of the random walk approach in the multi-label classification methods attracts many researchers’ sight. One challenge of using the random walk-based multi-label classification algorithms is to construct a random walk graph for the multi-label classification algorithms, which may lead to poor classification quality and high algorithm complexity. In this article, we propose a novel multi-label classification algorithm based on the random walk graph and the K-nearest neighbor algorithm (named MLRWKNN). This method constructs the vertices set of a random walk graph for the K-nearest neighbor training samples of certain test data and the edge set of correlations among labels of the training samples, thus considerably reducing the overhead of time and space. The proposed method improves the similarity measurement by differentiating and integrating the discrete and continuous features, which reflect the relationships between instances more accurately. A label predicted method is devised to reduce the subjectivity of the traditional threshold method. The experimental results with four metrics demonstrate that the proposed method outperforms the seven state-of-the-art multi-label classification algorithms in contrast and makes a significant improvement for multi-label classification.


Sign in / Sign up

Export Citation Format

Share Document