Arabic is one of the official languages recognized by the United Nations (UN) and is widely used in the middle east, and parts of Asia, Africa, and other countries. Social media activity currently dominates the textual communication on the Internet and potentially represents people’s views about specific issues. Opinion mining is an important task for understanding public opinion polarity towards an issue. Understanding public opinion leads to better decisions in many fields, such as public services and business. Language background plays a vital role in understanding opinion polarity. Variation is not only due to the vocabulary but also cultural background. The sentence is a time series signal; therefore, sequence gives a significant correlation to the meaning of the text. A recurrent neural network (RNN) is a variant of deep learning where the sequence is considered. Long short-term memory (LSTM) is an implementation of RNN with a particular gate to keep or ignore specific word signals during a sequence of inputs. Text is unstructured data, and it cannot be processed further by a machine unless an algorithm transforms the representation into a readable machine learning format as a vector of numerical values. Transformation algorithms range from the Term Frequency–Inverse Document Frequency (TF-IDF) transform to advanced word embedding. Word embedding methods include GloVe, word2vec, BERT, and fastText. This research experimented with those algorithms to perform vector transformation of the Arabic text dataset. This study implements and compares the GloVe and fastText word embedding algorithms and long short-term memory (LSTM) implemented in single-, double-, and triple-layer architectures. Finally, this research compares their accuracy for opinion mining on an Arabic dataset. It evaluates the proposed algorithm with the ASAD dataset of 55,000 annotated tweets in three classes. The dataset was augmented to achieve equal proportions of positive, negative, and neutral classes. According to the evaluation results, the triple-layer LSTM with fastText word embedding achieved the best testing accuracy, at 90.9%, surpassing all other experimental scenarios.

Automatic Text Summarization (ATS) is one of the utilizations of technological sophistication in terms of text processing assisting humans in producing a summary or key points of a document in large quantities. We use Indonesian language as objects because there are few resources in NLP research using Indonesian language. This paper utilized PLTMs (Pre-Trained Language Models) from the transformer architecture, namely T5 (Text-to-Text Transfer Transformer) which has been completed previously with a larger dataset. Evaluation in this study was measured through comparison of the ROUGE (Recall-Oriented Understudy for Gisting Evaluation) calculation results between the reference summary and the model summary. The experiments with the pre-trained t5-base model with fine tuning parameters of 220M for the Indonesian news dataset yielded relatively high ROUGE values, namely ROUGE-1 = 0.68, ROUGE-2 = 0.61, and ROUGE-L = 0.65. The evaluation value worked well, but the resulting model has not achieved satisfactory results because in terms of abstraction, the model did not work optimally. We also found several errors in the reference summary in the dataset used.

This study aims to apply the Internet of Things (IoT) technology to control electronic devices and monitor the electrical power usage remotely via the Internet. The system was implemented using Wemos D1, ACS712 current sensor, relay, and the Blynk application as the system interface on the smartphone. The system used an average of 0.4-3.3 seconds to respond to commands from the Blynk application via a Wifi connection at a distance of 50-1000 meters and the device control and power monitoring system can function properly. The system response time was not affected by distance. This system with Wifi access can be an alternative to control devices and monitor their power usage in addition to a longer time of SMS access and shorter range of Bluetooth.

Background: Hate speech is an expression to someone or a group of people that contain feelings of hate and/or anger at people or groups. On social media users are free to express themselves by writing harsh words and share them with a group of people so that it triggers separations and conflicts between groups. Currently, research has been conducted by several experts to detect hate speech in social media namely machine learning-based and lexicon-based, but the machine learning approach has a weakness namely the manual labelling process by an annotator in separating positive, negative or neutral opinions takes time long and tiringObjective: This study aims to produce a dictionary containing abusive words from local languages in Indonesia. Lexicon-base is very dependent on the language contained in dictionary words. Indonesia has thousands of tribes with 2500 local languages, and 80% of the population of Indonesia use local languages in communication, with the result that a significant challenge to detect hate speech of social media.Methods: Abusive words surveys are conducted by using proportionate stratified random sampling techniques in 4 major tribes on the island of Java, namely Betawi, Sundanese, Javanese, MadureseResults: The experimental results produce 250 abusive words dictionary from 4 major Indonesian tribes to detect hate speech in Indonesian social media by using the lexicon-based approach. Conclusion: A stratified random sampling technique has been conducted in 4 major Indonesian tribes to produce 250 abusive words for hate speech detection using the lexicon-based approach.

AbstrakKemajuan teknologi informasi berkembang secara pesat di berbagai bidang kehidupan. Internet adalah salah satu bagian dari teknologi informasi dan komunikasi mempunyai efek dan pengaruh yang sangat besar. Website salah satu teknologi internet tidak hanya sebagai media informasi tetapi menjadi proses pendukung bisnis perusahaan, akan tetapi penjualan melalui website belum cukup efektif jika tidak didukung dengan strategi promosi yang baik. SEO (Search Engine Optimization) adalah salah satu teknik promosi dengan cara memanfaatkan pengoptimalan mesin pencari agar website yang sudah kita buat berada diperingkat teratas atau halaman pertama (first page) sebuah halaman mesin pencari. Peneltian ini dilakukan pada sebuah website yang awalnya belum dilakukan teknik-teknik dari SEO, kemudian dengan menerapkan metode SEO on Page seperti optimasi keyword pada title tag, content, meta keyword,meta description, dan share ke sosial media, pada tahap ini juga dilakukan beberapa pengujian sebagai tolak ukur keberhasilan penerapan teknik-teknik SEO. Hasil dari penerapan teknik-teknik SEO mampu meningkatkan SERP (Search Engine Results Page) website di mesin pencari dan berhasil terindek oleh google berada di page kedua pada bulan kedua dan berhasil terindex di page pertama dalam pencarian google denga waktu kurang dari 3 bulan. Kata kunci : Website, SEO, Seo on Page, SERP AbstractThe advancement of information technology is growing rapidly in various fields of life. The internet is one part of information and communication technology has a very large effect and influence. The website not only serves as a medium of information but supports the company's business, but sales through the website are not effective enough if it is not supported by a good promotional strategy. SEO (Search Engine Optimization) is a promotional technique by utilizing search engine optimization so that the website we have created is at the top or first page (first page) of a search engine page.This research was conducted on a website that has not been done techniques from SEO, then by applying the SEO on Page method such as keyword optimization on title tags, content, meta keywords, meta description, and sharing to social media, at this stage several tests are also carried out. as a benchmark for the successful application of SEO techniques.The results of the application of SEO techniques are able to increase the SERP (Search Engine Results Page) of websites in search engines and successfully indexed by Google on the second page in the second month and successfully indexed on the first page in google search with less than 3 months. Keywords: Website, SEO, Seo on Page, SERP

It is challenging for humans to keep up with the rapid creation of digital information due to the explosion of digital information. A written document can be analyzed to extract meaningful information using automatic text summarization. This research proposes 16 different experimental settings in which the model developed by IndoBERT will be applied in order to answer the question of how much of an impact preprocessing has on the quality of summaries produced by automatic text summarization. In order to answer this question, the researchers have devised this study. In this study, we will explicitly talk about preprocessing strategies by conducting tests with different combinations of preprocessing techniques. These techniques include data cleansing, stopwords, stemming, and case folding. After that, the recall-oriented understudy for gisting evaluation (ROUGE) assessment will be used to conduct the measurement of the research results. According to the findings of this research, the optimal level of performance may be accomplished by combining the processes of data cleaning and case folding with scores of 0.78, 0.60, and 0.68 for ROUGE-1, ROUGE-2, and ROUGE-L respectively