Earticle

The original intention of this paper is to answer 柯理思question about the possible complements in Beijing dialect using only the form “V得C”. In order to solve this problem, we investigated the types and distribution of possible complements in northern dialects, which also involved the lexical form of “capability”. The category of “能性” is an important grammatical category. Chinese uses complements to indicate that the possible category is very special. A detailed description of the types and distribution of possible complements in Chinese is helpful for further research on the category of competence. The possible complements in northern dialects can be divided into three categories and four sub categories: dialects that use “V de(得) C“ (including “V de(得) C“) only, dialects using only “V (C) le(了)“, dialects using “V de(得) C“ (including “V de(得) le(了)“) and “VC le(了)“ (including “V le(了)“) ; The “V de(得) C“ in Beijing dialect is the result of the competition between “V de(得) C“ and “VC le(了)“; “Vle(了)lao(唠) → V lao(唠)“ is more likely that the grammatical meaning merges with the lexical meaning, rather than the “reduplicated syllable be omitted“.

This study demonstrates the influence of tech shock on consumer behavior. Amazon Go is a unique grocery store that enables people not to wait in line with their technologies. Therefore, Amazon Go can substitute the near restaurants. Collecting data from Yelp, we construct a panel dataset with various features of restaurants and implement Difference in differences estimation. We find (1) There is a negative impact on the local restaurant industry by the entry of Amazon Go. Specifically, consumers go to Amazon Go instead of the other restaurant mainly focusing on convenience but do not select Amazon Go as the alternative to fine dining. The other finding is (2) The standard of consumers is changed by Amazon Go. As Rebecca said (2016), people seem to feel traditional experiences burdensome after they experience new conveniences by technology. This approach enables business owners and retailers can get useful implications for their further strategies.

The primers for RT-PCR and RACE-PCR were designed by aligning the pig genomic sequence and the human complement factor B(CFB) coding sequence(CDS) from the GenBank. Each PCR product was amplified in pig cDNA and sequencing was carried out. The CDS length of pig CFB gene was determined to be 2298 bp. In addition, the pig CDS was more longer than human and mouse orthologs because of insertion and deletion. The identities of porcine nucleotide sequences with those of human and mice were 84% and 80%, and the identities of amino acids were 79% to 77%, respectively. Three complement control protein(CCP) domains, one Von Willebrand factor A(VWFA) domain and a serine protease domain, that are revealed typically in mammals, were found in the pig CFB gene. Based on the CDSs determined, the primers were designed in intron regions for amplification of entire length of exons. In amplification and direct sequencing with genomic DNAs of six pig breeds, three cSNPs(coding single nucleotide polymorphisms) were identified and verified as missense mutations. Using the Multiplex-ARMS method, we genotyped and verified the mutations identified from direct sequencing. To demonstrate recrudescence, we performed both direct sequencing and Multiplex-ARMS with two randomly selected DNA samples. The genotype of each sample exhibited the same results using both methods. Therefore, three cSNPs were identified from pig CFB gene and that can be used for haplotype analysis of the swine leukocyte antigen(SLA) class III region. Moreover, the results indicate that the Multiplex-ARMS method should be powerful for genotyping of genes in the SLA region.

The complementizer -ko in Korean before verbs of saying and thinking indicates the domain of speech and does not, have any modality. When it occurs with verbs of perception, it has a modality function, which indicates indirect evidence. In the so-called subject-to-object raising sentence, the raised nominal is not the grammatical subject of the complement but a previously mentioned discourse topic. By raising this topic, a speaker encodes a 'doubt in truth modality of the -ko complement clause.

The following are revealed regarding the correspondence relationship, which primarily exists between the presence of factivity in O case complements and the complementizers NO and KOTO in modern Japanese. a. The presence or absence of factivity in O case complements and the selection of NO or KOTO complementizers is not a one-to-one relationship; however, under certain conditions, a correlation exists between the allowance of the NO complement and complement factivity. b. The correlation is as follows: generally, higher tolerance of NO in a tensed complement allowing KOTO results in a stronger factive interpretation of the complement. Based on the behavior related to the above, it is therefore appropriate to separate the predicates into four types as follows. Type A :represented by “kateisuru(suppose)” Type B: represented by “noberu(state,claim)” Type C :represented by “tsutaeru(tell,report)” Type D: represented by “zannengaru(regret)” Type D, Type C, Type B and Type A bring the higher factive interpretation to complement proposition in that order, in parallel, NO in type D, Type C, Type B and Type A Indicates a higher acceptability in that order.

When the idiomatic theme forms an idiom chunk with the verb root in the multiple Case marking(MCM) construction, even if the Case marker isn’t attached to the theme, it may appear in the verbal complement position, preserving an idiomatic meaning. The idiomatic(i.e., special) meaning formed at the level of V’ may induce the awkward sentence to make complete sense. Thus, the idiomatic nominative theme in the embedded clause cannot occupy the embedded null subject position coindexed with the matrix ECMed element. The idiomatic nominative theme is in the verbal complement position of the embedded clause of the ECM construction, which is the reason why the idiomatic meaning stays the same, even though the sentence is negated. The head Neg cannot block the head movement at the level of V’ since Neg is between T and v(or V).(Lee, 2016b)

A Chinese complement construction composed of a head and its following complement contains distinctive linguistic features, while most other languages like Korean, English, etc. have no such features. The general classification of a complement construction by Chinese linguists is helpful but not enough to be completely transferred to a target language such as Korean. In this paper, we propose an approach where complement constructions are analyzed and classified based on the contrastive and computational viewpoint of the target language in Chinese‐to‐Korean machine translation. Further, the translation ambiguities of a head‐ndcomplement dependent case, which is the most complex type of complement construction in terms of translation, are resolved by k‐means clustering using mutual information. Some experimental results of such resolved ambiguities are provided from random samples. The proposed method improved the accuracy of disambiguation by 17.3 % with respect to the baseline.

본 연구에서는 언어가 틀린 믿음 수행을 예측해 주는지를 밝히기 위해 14명의 자폐 아동을 대상으로 실험을 실시하였다. 언어 능력은 마음상태 동사(생각하다)를 포함하는 보문, 의사소통 동사(말하다)를 포함하는 보문, 관계절 문장에 대한 이해를 측정하였고, 틀린 믿음 이해는 틀린 믿음 위치 과제와 내용 과제를 사용하여 측정하였다. 틀린 믿음 수행을 종속변인으로 삼고 마음상태 보문 이해, 의사소통 보문 이해, 관계절 이해를 예측 변인으로 하여 위계적 회귀분석을 실시하였다. 그 결과 마음상태 동사를 포함하는 보문에 대한 이해가 틀린 믿음 수행을 예측해 주었지만, 의사소통 동사를 포함하는 보문이나 관계절 문장에 대한 이해는 틀린 믿음 수행을 예측해 주지 못했다. 이러한 결과는 보문과 같은 특정한 구조에 대한 이해가 틀린 믿음 이해와 관련된다는 입장을 지지해준다 하겠다. 특히 마음상태를 표현하는 동사를 포함하는 보문만이 틀린 믿음 수행을 예측해 주었다는 점은 선행 연구 결과와 비교하여 논의할 필요가 있을 것이다. 또한 본 연구에 참여한 자폐 아동들의 틀린 믿음 과제에서의 수행은 정상 아동보다 저조하였는데, 이러한 결과는 자폐증의 마음이론 가설을 지지해 주는 것으로 해석될 수 있을 것이다.

This study examined the relationship between language and false belief in Korean children with autism (n=14). Specifically, it aimed to explore whether understanding of sentential complement structures could predict performance on false belief tasks in Korean children with autism. The children were tested with two kinds of false belief tasks (i.e., location change and content change tasks) and three kinds of language tasks (i.e., complement structure with a mental verb (think), complement structures with a communication verb (say), and relative clause structures). The tasks were presented as short stories with picture descriptions on a laptop computer. Questions were asked after each short story presentation. The data showed that the Korean children with autism performed lower than the typically developing Korean children did. Furthermore, multiple regression analyses indicated that the understanding of complement structures with a mental verb (think) predicted false belief task performance. But understanding of complement structures with a communication verb (say) and the relative clause comprehension were not predictive. The results support the view of de Villiers that false belief task performance is preceded by understanding of sentential complement structure. The lower performance of the Korean autistic children can also be interpreted as evidence that supports Baron-Cohen's theory of mind deficit hypothesis of autism.

본 연구는 소득인정액의 구성요소 중 하나인 주거용 재산을 소득으로 환산하는 방식이 갖는 정책적 타당성에 이의를 제기하고, 주거용 재산을 소득으로 환산하는 경우와 그렇지 않은 경우의 빈곤양태 변화를 살펴보았다. 이를 위하여 한국 복지패널(8차 자료)을 활용하여 추계한 주거재산 고려여부에 따른 가구별 소득인정액을 바탕으로 빈곤규모를 추정하였고, 주거재산의 소득환산이 초래하는 현상을 모의 분석하였다. 분석결과는 다음과 같다. 첫째, 주거재산을 소득으로 환산하는 경우 절대빈곤층(최저생계비의 100%)의 규모는 전체 가구의 8.7%인데 반해 주거재산을 소득으로 환산하지 않는 경우는 9.4%로 나타나 주거재산이 소득으로 환산되는 과정에서 전체 가구의 0.7%가 비수급 빈곤층으로 전락하는 것으로 나타났다. 둘째, 주거재산이 없는 빈곤위험층(소득인정액<최저생계비의 150%)의 경우, 필수비용인 주거비로 월평균 13.8만원을 단순 지출하고 있었고, 그 수준은 월평균 가처분소득의 17.2%을 차지하였다. 셋째, 주거재산의 소득환산과 같은 제도적 설계는 빈곤층의 탈빈곤을 저해하고 빈곤위험층의 방빈을 촉진하는 역할을 수행하는 것으로 판단된다. 이러한 분석결과를 토대로 본 연구는 주거재산의 소득환산방식이 빈곤층 혹은 빈곤위험층을 빈곤위험으로부터 보호하기 보다는 오히려 빈곤의 덫으로 유인할 가능성이 높음을 비판하였다. 일견 제도의 형평성을 강화하기 위한 재산의 소득환산제는 오히려 급여의 효율성을 악화시킬 뿐만 아니라 제도 본연의 목적 달성마저 방해하는 걸림돌이 되고 있다.

This paper object the validity of the method to calculate income converted amount of residential property, especially house, and examines the effect of that on a poverty phase. Main results are as follows; firstly, when residential property is converted into income, the poverty rate below minimum cost of living is 8.7%, but when residential property is not converted into income, the rate is 9.4%. Secondly, the possession of residential property increases the balance for living cost. Lastly, the design to calculate income converted amount of residential property would impede to emerge from poverty and also, accelerate to sink into poverty. As a result, this paper should underline the means test, which calculate income converted amount of residential property will worsen the efficiency of the allowance of public assistance and be an obstacle to achieve the purpose of public assistance.

본 연구는 현재 보편적으로 사용되고 있는 매체인 인터넷사용이 스마트폰이라는 신규 미디어가 도입될 때 매체이용의 대체‧보완 현상과 중독양상이 성인 및 청소년 세대별로 어떻게 나타나는지 살펴보았다. 연구대상은 청소년 집단은 수도권, 충청/강원, 호남(제주포함), 영남의 4개 지역 설문 응답자 3,816명 결과를 대상으로 하였으며 성인집단은 전문설문조사기관에 의뢰하였고 500명을 대상으로 하였다. 검사도구는 2011년 정보화진흥원에서 개발한 인터넷중독척도, 스마트폰중독척도와 정신건강척도를 포함하였다. 연구결과는 첫째, 인터넷중독 수준에 따라 스마트폰 중독수준의 차이를 카이검증 하였는데 인터넷 사용에 있어서는 문제가 없는 일반 사용자이지만 스마트폰은 중독적으로 사용하는 사람이 성인의 경우 58%, 청소년은 80%으로 청소년은 성인에 비해 스마트폰이 인터넷에 대체되어 사용되어 문제적 사용을 나타낼 수 있는 높은 경향성을 나타내었다. 둘째, 스마트폰의 주요용도에 있어서 청소년은 인터넷의 주활용 용도인 엔터테인먼트 기능이 스마트폰 사용에서도 두드러지게 나타나고 성인의 경우, 인터넷은 업무용, 스마트폰은 SNS 용도로 매체간 용도가 구분되어 사용되는 경향을 보인다. 셋째, 스마트폰 사용 이후 인터넷사용시간의 감소여부를 물었을 때 청소년은 줄었다는 응답이 66.1%, 성인은 44.8%로 나타나 청소년이 성인에 비해 스마트폰을 인터넷에 대체하는 매체로 사용할 수 있는 경향성을 뒷받침해 준다. 넷째, 인터넷과 스마트폰의 중독위험군을 대상으로 중독수준과 정신건강과의 관계를 살펴본 결과 청소년의 인터넷중독 수준과 정신건강의 수준만이 유의하게 상관이 있는 것으로 나타났는데 이는 성인에 비해 청소년의 인터넷중독이 정신건강에 미치는 위험성을 시사하는 부분이라 할 수 있다.

This study investigated how media addiction patterns are affected by the introduction of a new medium ? smartphone. In particular, we were interested in examining whether the new medium substituted or complemented the Internet in adults and youths. The participant pool consisted of 3,816 youths and 500 adults around Korea. Internet Addiction Proneness Scales, Smartphone Addiction Proneness Scales developed by the National Information Society Agency in 2011, along with the Mental Health Problems Scale, were used as measures. Our results showed that the percentage of normal Internet users in the high-risk smartphone addiction group was 58% for adults and 80% for youths. This suggests that youths are more likely to demonstrate problematic smartphone usage patterns after substituting smartphone for Internet. Second, youths mainly used their smartphones for entertainment purposes, just as they primarily used the Internet for the same reason. Adults, on the other hand, primarily used the Internet to do work, while they used their smartphones to get connected on SNS. Third, 66.1% of youths and 44.8% of adult participants replied that their smartphone usage reduced their time on the Internet. This reveals that youths are more likely to substitute smartphones for the Internet. Fourth, the correlations between addiction levels and mental health revealed that only the relationship between the youths’ Internet addiction level and their mental health was significant. This shows that the mental health of youths is more vulnerable to Internet addiction than that of adults.

This paper investigated complement coercion in English from computational psycholinguistic perspectives. In previous psycholinguistic behavioral studies, when humans processed sentences such as The author began the book, it took longer in reading time to process entity noun phrases (e.g., the book) after the next coercing verbs (e.g., begin) that require event arguments (e.g., writing the book) than the ones after non-coercing verbs (e.g., write) that do not. We leveraged one of the recent neural language models, GPT-2, to examine how it as a computational manifestation of the human mind engages in processing complement coercion in English. We found that GPT-2 yielded the higher surprisal value for coerced complements than for non-coerced complements, which is in keeping with the results from previous psycholinguistic behavioral studies. This finding shows that there is a close link between reading time in humans and the informational-theoretic measure of surprisal in GPT-2 during processing sentences containing complement coercion in English, and that in light of their correlation, GPT-2 seems to have learned relevant linguistic information bearing on complement coercion in English.