{ "paper_id": "O16-1020", "header": { "generated_with": "S2ORC 1.0.0", "date_generated": "2023-01-19T08:04:59.299148Z" }, "title": "Computing Sentiment Scores of Verb Phrases for Vietnamese", "authors": [ { "first": "Thien", "middle": [], "last": "Khai", "suffix": "", "affiliation": { "laboratory": "", "institution": "Ho Chi Minh City University of Technology Ho Chi Minh City", "location": { "country": "Vietnam" } }, "email": "thientk@cse.hcmut.edu.vn" }, { "first": "Tuoi", "middle": [ "Thi" ], "last": "Phan", "suffix": "", "affiliation": { "laboratory": "", "institution": "Ho Chi Minh City University of Technology Ho Chi Minh City", "location": { "country": "Vietnam" } }, "email": "" } ], "year": "", "venue": null, "identifiers": {}, "abstract": "Sentiment analysis is an emerging research field. One of the major tasks of sentiment analysis is building sentiment lexicons and calculating their scores, which is an essential job that provides \"material\" for all sentiment analysis problems. In this paper, we propose a fuzzy language computation by taking linguistic context into account to provide an effective method for computing the sentiment polarity of verb phrases. The positive results, which come from an experimental period, will provide us with a basis from which to build an effective sentiment analysis system by making use of the contextual valence shifter.", "pdf_parse": { "paper_id": "O16-1020", "_pdf_hash": "", "abstract": [ { "text": "Sentiment analysis is an emerging research field. One of the major tasks of sentiment analysis is building sentiment lexicons and calculating their scores, which is an essential job that provides \"material\" for all sentiment analysis problems. In this paper, we propose a fuzzy language computation by taking linguistic context into account to provide an effective method for computing the sentiment polarity of verb phrases. The positive results, which come from an experimental period, will provide us with a basis from which to build an effective sentiment analysis system by making use of the contextual valence shifter.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Abstract", "sec_num": null } ], "body_text": [ { "text": "Sentiment analysis (or opinion mining) is a new research field, but it is an important area that attracts the attention of not only researchers but also businesses and organizations. Building sentiment lexicons is an essential task that provides \"material\" for all sentiment analysis levels: document-based, sentence-based, concept-based, and aspect-based. One of the biggest English sentiment lexicons is SentiWordNet [15] . It contains opinion terms extracted from WordNet [3] with a semi-supervised learning method and is available for research purposes. SenticNet [2] is a lexical resource used in concept-level sentiment analysis. It provides sentiment scores for 14,000 common sense concepts. To tackle the problem of mining verb expressions to identify opinions from customer reviews, there also have been a large number of works discovered the semantics of verbs and verb phrases. For example, Sokolova and Lapalme [13] incorporate semantic verb categories including verb past and continuous forms into features sets. Neviarouskaya et al. [9] built a rule-based approach to incorporate verb classes from VerbNet [12] to detect the sentiment orientation of sentences.", "cite_spans": [ { "start": 419, "end": 423, "text": "[15]", "ref_id": "BIBREF14" }, { "start": 475, "end": 478, "text": "[3]", "ref_id": "BIBREF2" }, { "start": 568, "end": 571, "text": "[2]", "ref_id": "BIBREF1" }, { "start": 923, "end": 927, "text": "[13]", "ref_id": "BIBREF12" }, { "start": 1047, "end": 1050, "text": "[9]", "ref_id": "BIBREF8" }, { "start": 1120, "end": 1124, "text": "[12]", "ref_id": "BIBREF11" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1." }, { "text": "For Vietnamese, Vu et al. [18] built VietSentiWordNet, which contains 1,000 words; it also includes syntactic rules for extracting sentiments from review sentences. Hong et al. [4] built an opinion dictionary for product domains based on a combination of a statistical method, a machine translation technique, and WordNet. Their work outperformed VietSentiWordNet. Recently, in 2016, Son et al. [14] built a Vietnamese opinion dictionary that contains five sub-dictionaries: verb, adjective, adverb, noun, and proposed features. The sub-dictionaries are based on the English emotional analysis approach and adapted to traditional Vietnamese language. The support vector", "cite_spans": [ { "start": 26, "end": 30, "text": "[18]", "ref_id": "BIBREF17" }, { "start": 177, "end": 180, "text": "[4]", "ref_id": "BIBREF3" }, { "start": 395, "end": 399, "text": "[14]", "ref_id": "BIBREF13" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1." }, { "text": "The 2016 Conference on Computational Linguistics and Speech Processing ROCLING 2016, pp. 204-213 \uf0d3 The Association for Computational Linguistics and Chinese Language Processing machine classification technique was then used to identify the emotional content of the user's message. However, the authors calculated the sum of the emotional values of the linguistic variables based on feelings. In this paper, based on Vietnamese linguistic characteristics and the fuzzy computation proposed by Zadeh [6, 8, 19] , we present an effective method for computing the sentiment polarity of verb phrases. From this, we built a fine-grained linguistic sentiment analysis for Vietnamese. Zadeh developed the concept of fuzzy linguistic variables that modify the meaning and intensity of their operands, and we developed a modified fuzzy function suitable for use with the Vietnamese language. In our experiments, we showed that our system provides good results. In this paper, we describe our research contributions, as follows:", "cite_spans": [ { "start": 498, "end": 501, "text": "[6,", "ref_id": "BIBREF5" }, { "start": 502, "end": 504, "text": "8,", "ref_id": "BIBREF7" }, { "start": 505, "end": 508, "text": "19]", "ref_id": "BIBREF18" } ], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1." }, { "text": "-The mining of Vietnamese linguistic characteristics to propose sentiment computing rules for verb phrases.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1." }, { "text": "-Proposing the modified fuzzy functions suitable for Vietnamese linguistic variables.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1." }, { "text": "-Taking steps toward building an effective sentiment analysis system with fine-grained scores.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1." }, { "text": "The outline of the rest of this paper is as follows: in section 2, we present the linguistic characteristics of Vietnamese; in section 3, the proposed model is described; in section 4, we report our experiments; and finally, we conclude the paper and discuss possibilities for future work.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Introduction", "sec_num": "1." }, { "text": "Vietnamese is an isolating language with lexical tones and monosyllabic word structure. These characteristics are evident in all aspects: phonetic, vocabulary, and grammar. For vocabularies, Le [7] and Nguyen [11] proposed three common standards used to classify them: 1) essential meaning of the word type, 2) the function of the word in the sentence, and 3) the ability to combine with other words. Both Vietnamese and English words can be divided into content words and function words. Content words carry lexical meaning; while, function words relate lexical words to each other. For both languages, content words may be further divided into nouns, adjectives, and verbs.", "cite_spans": [ { "start": 194, "end": 197, "text": "[7]", "ref_id": "BIBREF6" }, { "start": 209, "end": 213, "text": "[11]", "ref_id": "BIBREF10" } ], "ref_spans": [], "eq_spans": [], "section": "Linguistic Characteristics of Vietnamese", "sec_num": "2." }, { "text": "Nouns are words that represent entities; adjectives represent qualities or characteristics; and verbs represent actions or states. In English, most adverbs are content words, but Vietnamese adverbs are function words. Generally, these words modify any part of speech other than a noun. Adverbs can modify verbs, adjectives, clauses, sentences, and other adverbs. In this paper, we only focus on verbs and adverbs.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Linguistic Characteristics of Vietnamese", "sec_num": "2." }, { "text": "Verbs denote action, state, or occurrence, and form the main part of the predicate of a sentence. In Vietnamese, there are some types of verbs [1, 10] ", "cite_spans": [ { "start": 143, "end": 146, "text": "[1,", "ref_id": "BIBREF0" }, { "start": 147, "end": 150, "text": "10]", "ref_id": "BIBREF9" } ], "ref_spans": [], "eq_spans": [], "section": "Vietnamese Verbs", "sec_num": "2.1" }, { "text": "Adverbs are words that modify or describe verbs, adjectives, clauses, sentences, and other adverbs. Generally, these words modify any part of speech other than a noun.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Vietnamese Adverbs", "sec_num": "2.2" }, { "text": "The following observations relate to Vietnamese adverbs when comparing them with English adverbs. Morphology. English adverbs are content words but Vietnamese adverbs are function words. To the best of our knowledge, there are approximately 600 Vietnamese adverbs while English has more than 6,000 adverbs. Syntactic. In English, the adverb is the head of the phrase, can appear alone, or can be modified by other words. An adverb phrase is a subordinate clause in a sentence. In Vietnamese, adverbs do not have primary grammatical functions in a clause (subject, predicate). Function. English adverbs modify a verb, adjective, or another adverb. The adverb typically expresses the manner, time, place, cause, or circumstance in which something has happened. In Vietnamese, adverbs do not have real meaning for describing the name, action, status, nature, and quantity of things. Adverbs only contain grammatical meaning based on the part of speech they modify. Position. There are three normal positions for adverbs in an English sentence: before the subject, between the subject and the verb, and at the end of the clause. Vietnamese adverbs can precede or follow the words they modify. Classification. English adverbs have the following types: time adverbs, degree adverbs, manner adverbs, frequency adverbs, and place adverbs.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Vietnamese Adverbs", "sec_num": "2.2" }, { "text": "For Vietnamese, a selection of the types of adverbs and their ability to combine with verbs are presented in Table 1 . ", "cite_spans": [], "ref_spans": [ { "start": 109, "end": 116, "text": "Table 1", "ref_id": "TABREF1" } ], "eq_spans": [], "section": "Vietnamese Adverbs", "sec_num": "2.2" }, { "text": "In this model, we try to compute the sentiment scores for word phrases that include verbs and adverbs based on Vietnamese linguistic characteristics. By combining with some adverbs, the verb phrases will have a smoother sentiment scaling.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Proposed Model", "sec_num": "3." }, { "text": "Our system architect is presented in Figure 1 . We used the English sentiment dictionary, SentiWordNet, and the translate tools Vdict * and Google Translate ** to build the core verb lexicons with sentiment scores for Vietnamese. The fuzzy rules then computed the sentiment scores for the whole phrase, which included the verbs and associated adverbs. Building core verb lexicons.", "cite_spans": [], "ref_spans": [ { "start": 37, "end": 45, "text": "Figure 1", "ref_id": "FIGREF0" } ], "eq_spans": [], "section": "System architecture", "sec_num": "3.1" }, { "text": "We constructed a handcrafted opinion dictionary containing approximately 1,000 verbs. The number of words was high enough to cater to the problem we sought to solve. These words:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "System architecture", "sec_num": "3.1" }, { "text": "appeared in the review corpus obtained from [16, 17] .", "cite_spans": [ { "start": 44, "end": 48, "text": "[16,", "ref_id": "BIBREF15" }, { "start": 49, "end": 52, "text": "17]", "ref_id": "BIBREF16" } ], "ref_spans": [], "eq_spans": [], "section": "System architecture", "sec_num": "3.1" }, { "text": "-are matched with corresponding English words in SentiWordNet; we used Vdict and Google Translate to check this. To meet the scope of this project, we assigned opinion word scores that were the same as the scores of words in SentiWordNet. In Table 2 , we describe some of the opinion words that appear in this core dictionary. ", "cite_spans": [], "ref_spans": [ { "start": 242, "end": 249, "text": "Table 2", "ref_id": "TABREF2" } ], "eq_spans": [], "section": "System architecture", "sec_num": "3.1" }, { "text": "Overall sentiment scores for the verb phrases were calculated thanks to fuzzy rules that were associated with the combination between the verb (denotes x) and the adverb (denotes y). We used fuzzy functions to incorporate the effect of the adverbs in the verb phrases. We considered the sentiment score of a verb to be its initial fuzzy score (x). Based on Vietnamese linguistic characteristics, we realized five sentiment shifting scalings for adverbs that go along with verbs; these were intensifier, booster, diminisher, minimizer, and modifier. General principles for classifying adverbs are as follows:", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "1. Adverbs of degree: There are five levels: intensifier, booster, diminisher, minimizer, and modifier. Some Vietnamese adverbs of degree are presented by Table 3. 2. Other adverbs: There are three levels that are booster, diminisher, and modifier:", "cite_spans": [], "ref_spans": [ { "start": 155, "end": 163, "text": "Table 3.", "ref_id": "TABREF3" } ], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "\uf0b7 Booster: PV1, PV2, PV31, PV41, PV6, PV9, PV10, PV13, PV16, PV17, PV18.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "\uf0b7 Diminisher: PV32, PV42, PV10, PV12, PV14, PV15.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "\uf0b7 Modifier: PH, PV19.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "Some of these adverbs are presented by Table 4 . In our system, Vietnamese adverbs are organized in a database. In Table 5 , we describe some of the adverbs that appear in our adverb database. In the table, \"Tag\" is the scaling category to which an adverb can belong. Similar to Zadeh's proposition [6, 8, 19] , if the verb phrase had an adverb, its modified fuzzy score was computed by (1):", "cite_spans": [ { "start": 299, "end": 302, "text": "[6,", "ref_id": "BIBREF5" }, { "start": 303, "end": 305, "text": "8,", "ref_id": "BIBREF7" }, { "start": 306, "end": 309, "text": "19]", "ref_id": "BIBREF18" } ], "ref_spans": [ { "start": 39, "end": 46, "text": "Table 4", "ref_id": "TABREF4" }, { "start": 115, "end": 122, "text": "Table 5", "ref_id": "TABREF5" } ], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "We chose = 4, 2, 1/2, or 1/4 if the adverb was a(n) intensifier, booster, diminisher, or minimizer.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "which gives us a modified fuzzy score, as indicated in (2) .", "cite_spans": [ { "start": 55, "end": 58, "text": "(2)", "ref_id": "BIBREF1" } ], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "with -\u0192(\u03bc(x),y) is the sentiment score of a verb phrase, in which x: verb, y: adverb.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "\u03bc(x) is the sentiment score of a verb. Table 6 presents an example of verb phrases and their sentiment scores. Table 6 . Sentiment score of verb phrases.", "cite_spans": [], "ref_spans": [ { "start": 39, "end": 46, "text": "Table 6", "ref_id": null }, { "start": 111, "end": 118, "text": "Table 6", "ref_id": null } ], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "According to the formula (2), if the adverb was a modifier (y.tag = modifier), we had two cases.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "For example: \u0192(tin ph\u1ea3i trust (a bad guy)) = -\u0192(tin trust) = -0.625, but \u0192(\u0111\u1eebng hi\u1ec3u l\u1ea7m shouldn't misconceive) = 0", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Fuzzy Rules", "sec_num": "3.2" }, { "text": "Cohen's kappa coefficient. Two judges participated in categorizing the adverbs as intensifier, booster, diminisher, minimizer, or modifier. To compute the \"between judges' agreement,\" we used the Cohen's kappa coefficient [5] , as follows: ", "cite_spans": [ { "start": 222, "end": 225, "text": "[5]", "ref_id": "BIBREF4" } ], "ref_spans": [], "eq_spans": [], "section": "Experiments", "sec_num": "4." }, { "text": "where Pr(a) is the relative observed agreement among the judges and Pr(e) is the hypothetical probability of a chance agreement. The Cohen's kappa coefficient of our corpus = 0.80. 104 Vietnamese verb phrases from Agoda.com were randomly collected to evaluate the system performance. The system was capable of handling 100 phrases. The highest sentiment score was +0.98 (c\u1ef1c k\u1ef3 tin t\u01b0\u1edfng extremely trust), and the lowest one was -0.99 (v\u00f4 c\u00f9ng gh\u00e9t extremely hate).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Experiments", "sec_num": "4." }, { "text": "Obviously, the adoption of fuzzy logic for computing sentiment scores of verb phrases helps the sentiment valences have a smoother sentiment scaling, not only 1, -1, and 0. In Table 7 , we describe the eleven levels of sentiment polarities that obtained from the testing. Table 7 . The eleven levels of sentiment polarities. Application. By identifying the fine-grained scores of phrase in sentence, the system can deal with many multi-class sentiment classification problems. For example, to classify the sentences, we simply counted the mean scores of sentiment phrases in each sentence. If the final score was more than +0.1 the sentence was considered to show a positive emotion. If the score was less than -0.1 the sentence was considered to show a negative emotion. Otherwise, the sentence was considered to show a neutral emotion.", "cite_spans": [], "ref_spans": [ { "start": 176, "end": 183, "text": "Table 7", "ref_id": null }, { "start": 272, "end": 279, "text": "Table 7", "ref_id": null } ], "eq_spans": [], "section": "Experiments", "sec_num": "4." }, { "text": "For example: R\u1ea5t tin t\u01b0\u1edfng v\u00e0o d\u1ecbch v\u1ee5 kh\u00e1ch s\u1ea1n, c\u1ef1c y\u00eau phong c\u1ea3nh n\u01a1i \u0111\u00e2y (Very trust in the hotel services, extremely love the scenery). Total score: (\u0192(r\u1ea5t tin t\u01b0\u1edfng very trust) + f(c\u1ef1c y\u00eau extremely love)) / 2 = (0.86 + 0.85) / 2 = 0.855. Therefore this sentence is considered to show a extremely positive emotion.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Experiments", "sec_num": "4." }, { "text": "This paper has presented a mechanism for computing the sentiment scores of verb phrases by mining the Vietnamese linguistic characteristics and using fuzzy functions. We have shown this approach to be effective. By identifying the opinion phrase polarity automatically, the method can be useful to deal with many sentiment analysis problems. Still, there are a number of challenges to indentify, classify, and calculate the sentiment scores of verbs and verb phrases because of linguistic challenges and the rule based approaches often suffer from domain-specificity problem. Future work will expand our research with more data and adopt this approach for developing Vietnamese sentiment lexicons with adjective phrases and noun phrases. We will also consider using machine learning methods to help the system become more robust.", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Conclusions", "sec_num": "5." }, { "text": "This paper was supported by the research project C2016-20-32 funded by Vietnam National University Ho Chi Minh City (VNU-HCM).", "cite_spans": [], "ref_spans": [], "eq_spans": [], "section": "Acknowledgment", "sec_num": "6." } ], "back_matter": [], "bib_entries": { "BIBREF0": { "ref_id": "b0", "title": "Vietnamese Grammar", "authors": [ { "first": "Hoang", "middle": [], "last": "Diep Van Ban", "suffix": "" }, { "first": "", "middle": [], "last": "Van Thung", "suffix": "" } ], "year": 1998, "venue": "Ng\u1eef ph\u00e1p ti\u1ebfng Vi\u1ec7t", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Diep Van Ban and Hoang Van Thung, (1998).: Ng\u1eef ph\u00e1p ti\u1ebfng Vi\u1ec7t, \"Vietnamese Grammar\", Vietnam Education Publishing House.", "links": null }, "BIBREF1": { "ref_id": "b1", "title": "SenticNet 3: A common and commonsense knowledge base for cognition-driven sentiment analysis", "authors": [ { "first": "Erik", "middle": [], "last": "Cambria", "suffix": "" }, { "first": "Daniel", "middle": [], "last": "Olsher", "suffix": "" }, { "first": "Dheeraj", "middle": [], "last": "Rajagopal", "suffix": "" } ], "year": 2014, "venue": "", "volume": "", "issue": "", "pages": "1515--1521", "other_ids": {}, "num": null, "urls": [], "raw_text": "Erik Cambria, Daniel Olsher, and Dheeraj Rajagopal, (2014). SenticNet 3: A common and common- sense knowledge base for cognition-driven sentiment analysis. AAAI, pp. 1515-1521.", "links": null }, "BIBREF2": { "ref_id": "b2", "title": "About WordNet", "authors": [], "year": 2010, "venue": "WordNet. Princeton University", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Princeton University \"About WordNet.\" WordNet. Princeton University. 2010. .", "links": null }, "BIBREF3": { "ref_id": "b3", "title": "Domain Specific Sentiment Dictionary for Opinion Mining of Vietnamese Text. The 8th Multi-Disciplinary International Workshop on Artificial Intelligence", "authors": [], "year": 2014, "venue": "", "volume": "", "issue": "", "pages": "136--148", "other_ids": {}, "num": null, "urls": [], "raw_text": "Hong Nam Nguyen, Thanh Van Le, Hai Son Le, Tran Vu Pham, (2014). Domain Specific Sentiment Dictionary for Opinion Mining of Vietnamese Text. The 8th Multi-Disciplinary International Workshop on Artificial Intelligence (MIWAI 2014): 136-148.", "links": null }, "BIBREF4": { "ref_id": "b4", "title": "Assessing agreement on classification tasks: the Kappa statistic", "authors": [ { "first": "J", "middle": [], "last": "Carletta", "suffix": "" } ], "year": 1996, "venue": "Computational Linguistics", "volume": "22", "issue": "", "pages": "249--254", "other_ids": {}, "num": null, "urls": [], "raw_text": "J. Carletta (1996). Assessing agreement on classification tasks: the Kappa statistic, Computational Linguistics, 22, 249-254.", "links": null }, "BIBREF5": { "ref_id": "b5", "title": "The concept of a linguistic variable and its application to approximate reasoning-II", "authors": [ { "first": "L", "middle": [ "A" ], "last": "Zadeh", "suffix": "" } ], "year": 1975, "venue": "Information Sciences", "volume": "8", "issue": "4", "pages": "301--357", "other_ids": {}, "num": null, "urls": [], "raw_text": "L. A. Zadeh, (1975). \"The concept of a linguistic variable and its application to approximate reasoning- II,\" Information Sciences,vol.8, no.4, part3, pp.301-357.", "links": null }, "BIBREF6": { "ref_id": "b6", "title": "Vietnamese Essentials: Grammar", "authors": [ { "first": "Le", "middle": [], "last": "Van Ly", "suffix": "" } ], "year": 1972, "venue": "S\u01a1 th\u1ea3o ng\u1eef ph\u00e1p Vi\u1ec7t Nam", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Le Van Ly, (1972) .:S\u01a1 th\u1ea3o ng\u1eef ph\u00e1p Vi\u1ec7t Nam, \"Vietnamese Essentials: Grammar\". Vietnam Education Publishing House.", "links": null }, "BIBREF7": { "ref_id": "b7", "title": "Opinion Mining from Online User Reviews Using Fuzzy Linguistic Hedges", "authors": [ { "first": "K", "middle": [], "last": "Mita", "suffix": "" }, { "first": "Mukesh", "middle": [ "A" ], "last": "Dalal", "suffix": "" }, { "first": "", "middle": [], "last": "Zaveri", "suffix": "" } ], "year": 2014, "venue": "Applied Computational Intelligence and Soft Computing", "volume": "2014", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Mita K. Dalal and Mukesh A. Zaveri, (2014). \"Opinion Mining from Online User Reviews Using Fuzzy Linguistic Hedges,\" Applied Computational Intelligence and Soft Computing. Volume 2014, Article ID 735942.", "links": null }, "BIBREF8": { "ref_id": "b8", "title": "Semantically distinct verb classes involved in sentiment analysis", "authors": [ { "first": "A", "middle": [], "last": "Neviarouskaya", "suffix": "" }, { "first": "H", "middle": [], "last": "Prendinger", "suffix": "" }, { "first": "M", "middle": [], "last": "Ishizuka", "suffix": "" } ], "year": 2009, "venue": "IADIS International Conference Applied Computing", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Neviarouskaya A., Prendinger H., Ishizuka M, (2009) \"Semantically distinct verb classes involved in sentiment analysis\", IADIS International Conference Applied Computing 2009.", "links": null }, "BIBREF9": { "ref_id": "b9", "title": "Vietnamese Grammar", "authors": [ { "first": "", "middle": [], "last": "Nguyen Kim Than", "suffix": "" } ], "year": 1997, "venue": "Nghi\u00ean c\u1ee9u ng\u1eef ph\u00e1p ti\u1ebfng Vi\u1ec7t", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Nguyen Kim Than, (1997) .: Nghi\u00ean c\u1ee9u ng\u1eef ph\u00e1p ti\u1ebfng Vi\u1ec7t, \"Vietnamese Grammar\". Vietnam Education Publishing House.", "links": null }, "BIBREF10": { "ref_id": "b10", "title": "Ng\u1eef ph\u00e1p ti\u1ebfng Vi\u1ec7t", "authors": [ { "first": "", "middle": [], "last": "Nguyen Tai Can", "suffix": "" } ], "year": 1975, "venue": "", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Nguyen Tai Can, (1975) .: Ng\u1eef ph\u00e1p ti\u1ebfng Vi\u1ec7t (Ti\u1ebfng -T\u1eeb gh\u00e9p -\u0110o\u1ea3n ng\u1eef), \"Vietnamese Grammar\". Hanoi Publisher.", "links": null }, "BIBREF11": { "ref_id": "b11", "title": "VerbNet overview, extensions, mappings and applications", "authors": [ { "first": "K", "middle": [ "K" ], "last": "Schuler", "suffix": "" }, { "first": "A", "middle": [], "last": "Korhonen", "suffix": "" }, { "first": "S", "middle": [ "W" ], "last": "Brown", "suffix": "" } ], "year": 2009, "venue": "HLT-NAACL", "volume": "", "issue": "", "pages": "13--14", "other_ids": {}, "num": null, "urls": [], "raw_text": "Schuler, K. K., Korhonen, A., and Brown, S. W. (2009). VerbNet overview, extensions, mappings and applications. In HLT-NAACL, 13-14.", "links": null }, "BIBREF12": { "ref_id": "b12", "title": "Verbs Speak Loud: Verb Categories in Learning Polarity and Strength of Opinions", "authors": [ { "first": "M", "middle": [], "last": "Sokolova", "suffix": "" }, { "first": "G", "middle": [], "last": "Lapalme", "suffix": "" } ], "year": 2008, "venue": "Canadian Conference on AI", "volume": "", "issue": "", "pages": "320--331", "other_ids": {}, "num": null, "urls": [], "raw_text": "Sokolova, M., and Lapalme, G. 2008. Verbs Speak Loud: Verb Categories in Learning Polarity and Strength of Opinions. In Canadian Conference on AI, 320-331.", "links": null }, "BIBREF13": { "ref_id": "b13", "title": "Recent Developments in Intelligent Information and Database Systems Volume 642 of the series Studies in Computational Intelligence", "authors": [ { "first": "Son", "middle": [], "last": "Trinh", "suffix": "" }, { "first": "Luu", "middle": [], "last": "Nguyen", "suffix": "" }, { "first": "Minh", "middle": [], "last": "Vo", "suffix": "" } ], "year": 2016, "venue": "", "volume": "", "issue": "", "pages": "263--276", "other_ids": {}, "num": null, "urls": [], "raw_text": "Son Trinh , Luu Nguyen, Minh Vo, Phuc Do, (2016) \"Lexicon-Based Sentiment Analysis of Facebook Comments in Vietnamese Language,\" Recent Developments in Intelligent Information and Database Systems Volume 642 of the series Studies in Computational Intelligence. pp 263-276.", "links": null }, "BIBREF14": { "ref_id": "b14", "title": "Sentiwordnet 3.0: An enhanced lexical resource for sentiment analysis and opinion mining", "authors": [ { "first": "Stefano", "middle": [], "last": "Baccianella", "suffix": "" }, { "first": "A", "middle": [ "E" ], "last": "Sebastiani", "suffix": "" }, { "first": "F", "middle": [], "last": "", "suffix": "" } ], "year": 2010, "venue": "LREC'10", "volume": "", "issue": "", "pages": "", "other_ids": {}, "num": null, "urls": [], "raw_text": "Stefano Baccianella, A.E., Sebastiani, F. (2010).: Sentiwordnet 3.0: An enhanced lexical resource for sentiment analysis and opinion mining. In: LREC'10 (May 2010).", "links": null }, "BIBREF15": { "ref_id": "b15", "title": "An upgrading SentiVoice -a system for querying hotel service reviews via phone", "authors": [], "year": 2015, "venue": "Proceedings of the 19th International Conference on Asian Language Processing (IALP 2015)", "volume": "", "issue": "", "pages": "115--118", "other_ids": {}, "num": null, "urls": [], "raw_text": "Thien Khai Tran and Tuoi Thi Phan, (2015) \"An upgrading SentiVoice -a system for querying hotel service reviews via phone,\" Proceedings of the 19th International Conference on Asian Language Processing (IALP 2015), Suzhou, China. pp. 115-118.", "links": null }, "BIBREF16": { "ref_id": "b16", "title": "Constructing Sentiment Ontology for Vietnamese Reviews", "authors": [], "year": 2015, "venue": "Proceedings of the 17th International Conference on Information Integration and Web-based Applications & Services (iiWAS2015)", "volume": "", "issue": "", "pages": "978--979", "other_ids": {}, "num": null, "urls": [], "raw_text": "Thien Khai Tran and Tuoi Thi Phan, (2015) \"Constructing Sentiment Ontology for Vietnamese Reviews,\" Proceedings of the 17th International Conference on Information Integration and Web-based Applications & Services (iiWAS2015), Brussels, Belgium. December 11 -13, 2015. pp.281-285. ISBN: 978-1-4503-3491.", "links": null }, "BIBREF17": { "ref_id": "b17", "title": "A feature-based opinion mining model on product reviews in Vietnamese", "authors": [ { "first": "Tien-Thanh", "middle": [], "last": "Vu", "suffix": "" }, { "first": "Huyen-Trang", "middle": [], "last": "Pham", "suffix": "" }, { "first": "Cong-To", "middle": [], "last": "Luu", "suffix": "" }, { "first": "Quang-Thuy", "middle": [], "last": "Ha", "suffix": "" } ], "year": 2011, "venue": "Semantic Methods for Knowledge Discovery and Communication, Polish-Taiwanese Workshop", "volume": "", "issue": "", "pages": "22--23", "other_ids": {}, "num": null, "urls": [], "raw_text": "Tien-Thanh Vu, Huyen-Trang Pham, Cong-To Luu, Quang-Thuy Ha (2011).: A feature-based opinion mining model on product reviews in Vietnamese. In: Semantic Methods for Knowledge Discovery and Communication, Polish-Taiwanese Workshop, Springer Berlin Heidelberg. pp.22-23.", "links": null }, "BIBREF18": { "ref_id": "b18", "title": "A parametric representation of linguistic hedges in Zadeh's fuzzy logic", "authors": [ { "first": "V", "middle": [ "N" ], "last": "Huynh", "suffix": "" }, { "first": "T", "middle": [ "B" ], "last": "Ho", "suffix": "" }, { "first": "Y", "middle": [], "last": "Nakamori", "suffix": "" } ], "year": 2002, "venue": "International Journal of Approximate Reasoning", "volume": "30", "issue": "3", "pages": "203--223", "other_ids": {}, "num": null, "urls": [], "raw_text": "V. N. Huynh, T. B. Ho, and Y. Nakamori, (2002) \"A parametric representation of linguistic hedges in Zadeh's fuzzy logic,\" International Journal of Approximate Reasoning, Vol. 30, No. 3, pp.203-223.", "links": null } }, "ref_entries": { "FIGREF0": { "num": null, "uris": null, "type_str": "figure", "text": "http://vdict.com ** https://translate.google.com/ System architecture." }, "TABREF0": { "content": "", "type_str": "table", "html": null, "num": null, "text": "Transitive verbs are action verb that have an object to receive that action. For example: l\u00e0m do, tr\u1ed3ng plant, x\u00e2y build, ph\u00e1t tri\u1ec3n develop, \u0111\u00e0n \u00e1p suppress, mua b\u00e1n purchase etc. -Verb of giving and receiving (denotes Vex2): For example: cho give, g\u1eedi send, t\u1eb7ng offer, bi\u1ebfu donate etc.; nh\u1eadn get, vay lend etc. -Verb of command (denotes Vex3): this type of verb presents activities that promote or prevent one from doing something else. For example: khuy\u00ean advice, b\u1eaft bu\u1ed9c obligatory, \u0111\u1ec1 ngh\u1ecb suggest, \u0111\u00ecnh ch\u1ec9 suspend etc. -Verb of moving, direction (denotes Vdr). For example: v\u00e0o in, ra out, l\u00ean up, xu\u1ed1ng down, \u0111\u1ebfn come, l\u1ea1i back etc. Verb of mentality, awareness (denotes Vin1): h\u1ed1i ti\u1ebfc regret etc. -Verb of emotion (denotes Vin2): h\u1ea1nh ph\u00fac happy, bu\u1ed3n sad, gi\u1eadn angry etc. -Verb of physiology (denotes Vin3): mong want etc. -Verb of nature, morality, personality (denotes Vin4): nh\u1ecbn condescend, tha th\u1ee9 forgive etc." }, "TABREF1": { "content": "
TypesAdverbsKinds of verbsVerb phrases
PV1 the same, similar\u0111\u1ec1u both c\u0169ng too c\u00f9ng jointlyPV1 + (Vdr, Vin1, Vin2, Vin3, Vin4, Vex1, Vex2, Vex3)c\u00f9ng chu\u1ea9n b\u1ecb (prepared jointly)
PV2 continuationv\u1eabn stillPV2 + (Vdr, Vin1, Vin2, Vin3, Vin4, Vex1, Vex2, Vex3)v\u1eabn c\u01b0\u1eddi (still smile)
PV31 time relation (present+s\u1ebd will \u0111ang -ingPV3 + (Vdr, Vin1, Vin2, Vin3, Vin4, Vex1, Vex2, Vex3)Anh s\u1ebd thi r\u1edbt. (He will fail the exam.)
future)
PV32 time relation (pass)v\u1eeba just \u0111\u00e3 -ed t\u1eebngPV3 + (Vdr, Vin1, Vin2, Vin3, Vin4, Vex1, Vex2, Vex3)Anh \u1ea5y t\u1eebng thi r\u1edbt. (He has failed the exam.)
PV41 frequency (increase)th\u01b0\u1eddng usually hay often n\u0103ng alwaysPV4 + (Vdr, Vin1, Vin2, Vin3, Vin4, Vex1, Vex2, Vex3)hay \u0103n tr\u1ec5 (often eat lately)
PV42 frequency (decrease)\u00edt rarely hi\u1ebfm rarelyPV4 + (Vdr, Vin1, Vin2, Vin3, Vin4, Vex1, Vex2, Vex3)\u00edt \u0111i tr\u1ec5 (rarely go late)
PV5 degreer\u1ea5t very, h\u01a1i a bit qu\u00e1 too, l\u1eafm much c\u1ef1c extremelyPV5 + (Vin1, Vin2, Vin3)r\u1ea5t y\u00eau (very love)
PV6 confirmationc\u00f3 toPV6 + (Vdr, Vin1, Vin2, Vin3, Vin4, Vex1, Vex2, Vex3)c\u00f3 t\u1ed3n t\u1ea1i (to exist)
PV7\u0111\u1eebng don'tPV7 + (Vdr, Vin1, Vin2, Vin3,ch\u1edb hi\u1ec3u l\u1ea7m (shouldn't
commandch\u1edb shouldn'tVin4, Vex1, Vex2, Vex3)misconceive)
PHkh\u00f4ng don't ch\u01b0aPH + (Vdr, Vin1, Vin2, Vin3,kh\u00f4ng \u0111i
negationyetVin4, Vex1, Vex2, Vex3)(don't go)
", "type_str": "table", "html": null, "num": null, "text": "Vietnamese adverbs and their ability to combine with verbs." }, "TABREF2": { "content": "
TermPositive ScoreNegative ScorePOSTag
y\u00eau love0.3750VerbVin2
gh\u00e9t hate00.75VerbVin2
tin t\u01b0\u1edfng trust0.6250VerbVin2
k\u00ednh n\u1ec3 respect0.50VerbVin2
", "type_str": "table", "html": null, "num": null, "text": "Fragment of Core Opinion Dictionary." }, "TABREF3": { "content": "
intensifierboosterdiminisherminimizermodifier
c\u1ef1c k\u1ef3 extremelyr\u1ea5t verykh\u00e1 ratherc\u0169ng seeminglykh\u00f4ng no
c\u1ef1c stronglyqu\u00e1 toot\u01b0\u01a1ng \u0111\u1ed1i relativelyh\u01a1i a bitch\u1eb3ng no
si\u00eau superl\u1eafm mucht\u1ea1m ratherr\u1ed3i alreadych\u1ea3 no
", "type_str": "table", "html": null, "num": null, "text": "Some Vietnamese adverbs of degree with their scalings." }, "TABREF4": { "content": "
boosterdiminishermodifier
\u0111\u1ec1u bothph\u1ea3i toch\u1ea3 no
v\u1eabn stillhi\u1ebfm rarelykh\u00f4ng not
hay oftent\u1eebng alreadych\u01b0a yet
", "type_str": "table", "html": null, "num": null, "text": "Some other adverbs with their scalings." }, "TABREF5": { "content": "
AdverbsTypesTag
c\u1ef1c k\u1ef3 extremelyPV5intensifier
kh\u00f4ng noPHmodifier
ph\u1ea3i toPV19modifier
hay oftenPV41booster
hi\u1ebfm rarelyPV42diminisher
", "type_str": "table", "html": null, "num": null, "text": "Some Vietnamese adverbs with their tags." } } } }