Hyper-dense wireless network deployment is one of the popular solutions to meeting high capacity requirement for 5G delivery. However, current operator understanding of consumer satisfaction
comes from call centers and base station quality-of-service (QoS) reports with poor geographic accuracy. The dramatic increase in geo-tagged social media posts adds a new potential to understand consumer satisfaction towards target-specific quality-of-experience (QoE) topics. In our paper, we focus on evaluating users’ opinion on wireless service-related topics by applying natural language processing (NLP) to geo-tagged Twitter data. Current generalized sentiment detection methods with generalized NLP corpora are not topic specific. Here, we develop a novel wireless service topic-specific sentiment framework, yielding higher targeting accuracy than generalized NLP frameworks. To do so, we first annotate a new sentiment corpus called SignalSentiWord (SSW) and compare its performance with two other popular corpus libraries, AFINN and SentiWordNet. We then apply three established machine learning methods, namely: Naive Bayes (NB), Support Vector Machine (SVM), and Recurrent Neural Network (RNN) to build our topic-specific sentiment classifier. Furthermore, we discuss the capability of SSW to filter noisy and high-frequency irrelevant words to improve the performance of machine learning algorithms. Finally, the real-world testing results show that our proposed SSW improves the performance of NLP significantly.

/pdf/mapping-consumer-sentiment-toward-wireless-services-using-1h5q35xhq1.pdf

Mapping consumer sentiment toward wireless services using geospatial twitter data

CyberAid: Are your children safe from cyberbullying?

In the increasingly fierce restaurant business competition, many restaurants compete to provide the best quality for consumers. The quality of the restaurant includes food and drink, environment, place and service. The condition affects the restaurant brand image that is marked whether consumers are satisfied or not. One of review restaurant website called TripAdvisor is chosen as the data source because it contains User-Generated Content features (UGC). The advantage of using UGC is to ensure the authenticity of consumer comment data. English texts classification is used in this study to determine dissatisfaction (negative) and satisfaction (positive) of consumers based on their comment or review. This paper utilises Decision Tree-J48 as the classification method in this study. Data retrieval is done by crawling data using WebHarvy. The overall performance of the Decision Tree-J48 method is the average value of Precision 48.7%, Recall 36.8%, F-Measure 41.40% and accuracy 45.6%. The benefit of this classification results is as a recommendation for consumers to choose the best restaurant.

Sentiment Analysis of Restaurant Review with Classification Approach in the Decision Tree-J48 Algorithm

Social media has grown so rapidly, so people easily to share their opinions, moments, etc. There are several types of research about social media, one of which is Sentiment Analysis (SA) that can also be referred to as opinions meaning (OM). Sentiment Analysis focuses on the classification of patterns that are derived from words that are positive words, negative words, and neutral words. In this paper, the researcher uses sentiment analysis with a machine learning approach and uses Fuzzy K-Nearest Neighbor (FK-NN) as the classification method. The dataset uses English text classification, to predicted sentiment of customer reviews about the positive or negative review. The predicted results show that Sentiment Analysis FK-NN is slightly close to the results of the previous research method, namely Probabilistic Latent Semantic Analysis (PLSA), which FK-NN is 72.05% and PLSA is 76%.

Fuzzy K-Nearest Neighbor for Restaurants Business Sentiment Analysis on TripAdvisor

Twitter is a powerful social media where people share their opinion on various topics. Sentiment Analysis on twitter data gives the classification of opinion on a topic as positive, negative or neutral. Twitter messages are written informally and tweets are short. Hence, the classification of tweets by only considering the text part of the message does not give accurate results. To improve the classification accuracy we use Emotion Tokens like Emoticons or Emojis. Emotion Tokens are independent of language, grammar or size of the tweet. Considering Emotion tokens while classifying tweets will improve the accuracy of classification. In this paper, we propose unsupervised Sentiment classification on Twitter Data using Emoticons to improve the performance of classification.

Unsupervised Sentiment Classification of Twitter Data using Emoticons

The large number of internet users caused increasing the number of social media users. Twitter is one of social media that have a large number of users in Indonesia. As a social media, twitter allows users to share information via status in a tweet. Due to the limitations of the use of text is only 280 characters, emoticons are commonly used in tweet. Emoticon can explain the condition or feeling which is described in a text-shaped punctuation mark. This paper will focus on creating emoticon dictionary and weighting of an emoticon. Emoticon dictionary contains a list of 384 emoticons describing a variety of feelings and emotions. The used dataset contains Indonesian language tweets from twitter API. We tried to analyze sentiment on existing datasets with reference scores in SentiWordNet. Weighting emoticons done under the assumption that the emoticons have more effect in a sentence than ordinary words. After that, we classify the results into three classes, namely sentiment positive, negative and neutral. We compared the results between the emoticon-based algorithm and without considering emoticons algorithm. Accuracy obtained on the emoticon-based using algorithm is 0.74.

Riyananto Sarno

Papers

Sentiment Analysis Using Weighted Emoticons and SentiWordNet for Indonesian Language