Previous research Sample Clauses

Previous research. Sarikoli is an underdescribed and poorly documented language. Arlund de- scribes it as “the most isolated and understudied of the [Pamir] languages” as a result of its confinement to a remote border area of China, presenting great challenges to linguists in terms of geographical remoteness, requirement of Mandarin proficiency, and the red tape and surveillance of the Chinese gov- ernment (Arlund 2006:6). ▇▇▇▇▇▇▇▇ speculates that ▇▇▇▇▇▇▇▇ has kept many words and forms lost in other Pamir languages due to its geographical and political isolation from other Pamir languages (Paxalina 1966:4). Few linguists have produced descriptions of Sarikoli based on data from their own fieldwork, and they will be introduced in this section. Although ▇▇▇▇▇▇▇▇ has also been mentioned in several general works on Pamir languages or the Shughni-Rushani subgroup (▇▇▇▇▇ 1933; ▇▇▇▇▇ 1936; Morgenstierne 1938 & 1974; ▇▇▇▇▇ 1989; ▇▇▇▇▇▇▇ 1989; ▇▇▇▇▇▇▇ & Dodykhudoeva 2009a; ▇▇▇▇▇- land 2009), those works are based on materials published by those who did original research in the 1870s and 1950s: ▇▇▇▇ (1876) and ▇▇▇▇▇▇▇▇ (1966). The first English mention of ▇▇▇▇▇▇▇▇ appeared in 1875, when Britain sent an official mission to Eastern Turkestan (present-day Xinjiang) led by diplomat ▇▇▇▇▇▇ ▇▇▇▇▇▇▇ ▇▇▇▇▇▇▇ in 1873, during the closing decades of the Great Game, the struggle between Victorian Britain and Tsarist Russia for geopoliti- cal power in Central Asia. Two of the participants of this expedition, medical ▇▇. ▇▇▇▇▇ ▇▇▇▇▇▇ ▇▇▇▇▇▇ and Colonel ▇▇▇▇ ▇▇▇▇▇▇▇▇, collected substantial wordlists and twenty phrases of Sarikoli (to which they refer as ▇▇▇▇▇▇ ▇▇▇▇ and Sirikolee, respectively). These data are in chapter 15 of Forsyth’s report on this mission, which also includes rich historical, geographical and ethnograph- ical information on western Xinjiang (Forsyth 1875). ▇▇▇▇▇▇ and Biddulph’s wordlists can be useful for historical-comparative work. The first English description of Sarikoli was written by ▇▇▇▇▇▇ ▇. ▇▇▇▇, a British political agent who was on special duty at Kashgar (▇▇▇▇ 1876). In 1868, he was “the first Englishman who ever went to Yarkund” (Forsyth 1871), a county off the northeast border of Varshide, just a short distance away from the village of Teeng. In 1872, when he returned to England, he was awarded the patron’s gold medal by the Royal Geographical Society for his service in exploring Eastern Turkestan (▇▇▇ 1897). He also published several linguis- tic descriptions of the...

Previous research. Rigorous description of rGyalrong languages starts among Chinese linguists in the 1940s with ▇▇▇ ▇▇▇▇ (Kin 1949). A large-scale linguistic survey of rGyalrongic was conducted in the 1950s, and some of the results are published in ▇▇▇ ▇▇▇▇▇▇▇▇▇’s magnum opus Lin (1993), along with his own research. More recently, important work was done by ▇▇▇▇▇▇▇ ▇.-S. ▇▇▇, his student ▇▇▇ ▇▇▇-▇▇▇▇, and ▇▇▇▇▇▇▇▇▇ ▇▇▇▇▇▇▇, with his book (▇▇▇▇▇▇▇ 2008) on all the rGyalrong group, concentrating on the Japhug of gDongbrgyad. Zbu rGyalrong is the least documented language in the group. No text has been published in this language. ▇▇▇ (1993) recorded the vocabulary of the dialect of the village of Zhongre. ▇▇▇▇▇▇▇

Previous research. Although there is limited prior work addressing NLP and IR in the archaeology domain, there are some examples of related research in the literature. Almost all of those studies have focused on grey literature as the source material, presumably because it has the greatest potential for computational techniques. One of the earliest applications of IR in archaeology was done by ▇▇▇▇▇▇▇▇ (1983), who did a study on information needs of users of a sites and monuments record. As this was back in 1983, the information was stored on physical 5 by 8 inch record cards, ordered by grid coordinates. Even though the situation was very different to our current situation, the problem is the same: the metadata (grid coordinates) were not good enough for information retrieval, as users want a way to cross reference or search through the data (text on cards). ▇▇▇▇▇▇▇▇ sent out surveys by post asking archaeology professionals on their opinion on the use of computers for record manipulation, and found that 63% already did, or were hoping to do so in the future, meaning 37% of respondents did not see any value in using computers for this task. Eventually they concluded that “A computer-based recording system gives the potential to relieve problems of lack of space, lost data, inaccuracies in recording and to provide a flexible and efficient retrieval system, therefore relieving staff time for other work” (▇▇▇▇▇▇▇▇, 1983, p. 43), which is basically also the main aim of this project. It seems not much has changed in the last 40 years in that respect. At the end of the 20th century, computer systems became increasingly com- mon place, and in the last 20 years a number of projects have used Text Mining techniques on archaeological texts. ▇▇▇▇▇▇ et al. (2008) created a full work- flow allowing experts to extract information from text, but in a quite specialised way on small collections, and is not meant for searching through large corpora. ▇▇▇▇▇ & ▇▇▇▇▇ (2010) experimented with extracting archaeological events and converting them to Resource Description Framework (RDF) triples, to increase the interconnectivity between data silos. Going more in the direction of IR, the Archaeotools project used a combina- tion of rules based and machine learning approaches to automatically generate location, time period, and subject metadata for a small selection of a thousand reports, with moderate success. This generated metadata could then be used for searching in a facetted interface (▇▇▇▇▇▇▇ et...

Previous research. 31 algorithm to perform NER. This showed promising results, but unfortunately the technique has not been evaluated fully yet. Building on her work, Talks (2019) added more entity types and did an extensive evaluation with users. All the research described above has been on the English language, and re- search on Dutch and other languages is much less prevalent. For Dutch, there are two main examples: the OpenBoek project and the experiments on Dutch texts in the above mentioned ARIADNE project. The OpenBoek project (▇▇▇▇▇▇▇▇ & ▇▇▇▇▇▇, 2008; ▇▇▇▇▇▇▇▇ & Brandsen, 2010) aimed to create a full text search engine combined with entity search, on about 2,000 reports. They used Memory Based Learning to automatically label time periods and locations, which were searchable together with the full text in a web application based on the SMART system (Salton, 1971). While the search engine showed promising results, unfortunately this web application has gone offline not too long after the funding for the project ended. The ARIADNE project – besides the work on English texts described above – also experimented with Dutch and Swedish grey literature. For Dutch, they applied a rules based technique using the General Architecture for Text Engineer- ing (GATE) framework (▇▇▇▇▇▇▇▇▇▇ et al., 1995). The rules were mainly based on thesauri, but they found many issues with the thesauri and gold standard, making effective NER with this approach difficult. Very recently, ▇▇▇▇▇▇▇ et al. (2021) experimented with Text Mining and IR as part of their research on urban farming and ruralisation in the Netherlands. They extracted text from a number of PDFs, created a term document matrix and compared this with a list of keywords related to the topic of urban farming, to automatically assess the relevance of a large number of documents for a number of topics. In a slightly different direction, recent work by ▇▇▇▇▇ et al. describes research on Dutch archaeological texts from Belgium, looking at theoretical trends over time. They successfully manage to use Text Mining to find these trends, and chart the decrease in text quality due to developer-led archaeology. Similarly, ▇▇▇▇▇▇▇ et al. (2020) used topic modelling techniques on large-scale English data to see if there are patterned ways in which archaeologists write about bone. Almost no research has been done on multilingual techniques, but ▇▇▇▇▇▇▇- ▇▇▇▇▇▇▇ et al. (2015) present some interesting results for NER on English, German and French ...

Previous research. In this section I talk about some articles I have found interesting for my research study during my review of previously research in this area. In general, there are many studies about how we can improve our safety procedures practices, and I have found two main approaches in this area: researchers, trying to create and simulate their own environment. Hopefully, I do not need to create new one, because I will use Second Life as my environment. Secondly, researchers were focusing on procedures and advantages or disadvantages applying these procedures in virtual world. But, none of them were using Second Life yet. Let me introduce some of the articles.

Previous research. In the other fields of the social sciences, the previous research concerning immigration from former Soviet republics has concentrated on social networks, stigmatization experiences, integration, professional experiences and migration of mothers of young children. Therefore, this field has been broadly studied already. The viewpoint of the most of the previous studies made in Finland, has been either to study the number of immigrants and to clarify their backgrounds by using migration ▇▇▇▇▇▇▇▇▇ or to research the acculturation process of the immigrants8. Also studies concerning the attitudes of Finns against foreigners9, the willingness of Estonian people to emigrate10 and Estonian people's stigmatization experiences11 have been made. In the field of cross-cultural psychology, for example ▇▇▇▇ ▇. ▇▇▇▇▇ has studied the psychological consequences of acculturation and created concepts to study acculturation and adaptation.12 Finnish immigration research has traditionally concentrated on how immigrants have done and acted in Finland and among Finns but less attention has been paid on the communities 7 E.g. Kyntäjä & Kulu 1998. 8 E.g. Liebkind 1994, Perhoniemi & ▇▇▇▇▇▇▇▇▇▇-Lahti 2006, ▇▇▇▇▇▇▇▇▇▇ 2006.

Previous research. Background research for the archaeological survey of the US 301 project began with the development of a predictive model for the occurrence of prehistoric and historic archaeological resources within a broad study area that encompassed several proposed highway alignments (▇▇▇▇▇▇▇▇ et al. 2006). For the occurrence of prehistoric archaeological resources, the model relied primarily on the cost distance to various sources of water and the presence of micro- drainage divides. For the occurrence of historic resources, the model was based on the locations of extant historic structures, structures shown on historic maps, and proximity to early roads. This was followed by a Phase Ia survey for US 301 Section 1, in which the predictive model was evaluated and applied specifically to the project area (Hay et al. 2009). This resulted in the division of the US 301 Section 1 project area into 31 survey segments, based on zones of high, medium, or low probability for prehistoric and historic archaeological resources (see Figures 2a-2f). Background and archival research was conducted in conjunction with the Phase Ia survey, to trace the history of ownership of the historic properties within the project area. At that time, only two of the 12 identified historic properties through which Section 1 passes were successfully traced back to original land grants. The others were generally traced back to mid- nineteenth century ownerships, with the aid of names shown on the 1849 and 1868 atlases (Rea and Price 1849, Beers 1868), but could not be traced further back for various reasons.

Previous research. Research has been done on whether the drawings in the Paris manuscript are made from life and how to interpret the claim of “na het leven.” When this manuscript was exhibited in Middelburg in 1987, Blussé and ▇. ▇▇▇▇▇▇▇▇▇▇ each wrote an essay for the catalogue ▇▇▇▇▇ ▇▇▇▇▇▇▇▇ Beelden van en Chinareis 1655–1657, and their investigations had a profound impact on the later study of ▇▇▇▇▇▇▇’s work of China. ▇▇▇▇▇▇’s essay concentrates on introducing of the historical background of the VOC envoys’ first visit to China, the publication of ▇▇▇▇▇▇▇’s book and its different editions, and particularly the discovery of the Paris manuscript, which he argues is the source of ▇▇▇▇▇▇▇’s book. In his opinion, ▇▇▇▇▇▇▇ himself compiled the manuscript on the basis of the sketches he made on spot, and the lack of trustworthiness of various published editions should be attributed to arbitrary changes made by the editors and publishers.55 This has laid an essential foundation for the later study. Writing from the perspective of an art historian, ▇▇▇▇▇▇▇▇▇▇ focuses more on the examination on the trustworthiness of the drawings in the Paris manuscript and pointed out the importance of the European audience’s understanding of the phase “na het leven” in the seventeenth century. He agrees with the historian ▇. ▇▇ ▇▇▇▇’▇ opinion that the engravings in ▇▇▇▇▇▇▇’s book of China were made on the basis of the sketches that ▇▇▇▇▇▇▇ made in China, but that they were later supplemented with cityscapes in the background and figures in the foreground by other people in the Netherlands. To support this hypothesis, he has analysed a number of drawings to determine how they were produced. On the basis of his findings, he argues that the drawings in the Paris manuscript are not the original sketches that ▇▇▇▇▇▇▇ made on site but were made on the basis of some rough sketches. He believes the drawings in the Paris manuscript were made by different hands. He concludes that although these engravings supposedly represented “the more realistic China,” they actually reproduced exotic and 55 Blussé and ▇▇▇▇▇▇▇▇▇▇, ▇▇▇▇▇ ▇▇▇▇▇▇▇▇ Beelden Van Een Chinareis 1655–1657, 19–20. fantastic images of China.56 ▇▇▇▇▇▇▇▇▇▇’▇ research pays much attention to the drawings and engravings themselves and investigates them from different angles and pays special attention to how they were produced. It brings new perspectives to the study of the drawings, especially in respect to the “na het leven” quality, but to a certain d...

Previous research. ‌ Although there is limited prior work addressing NLP and IR in the archaeology domain, there are some examples of related research in the literature. Almost all of those studies have focused on grey literature as the source material, presumably because it has the greatest potential for computational techniques. One of the earliest applications of IR in archaeology was done by ▇▇▇▇▇▇▇▇ (1983), who did a study on information needs of users of a sites and monuments record. As this was back in 1983, the information was stored on physical 5 by 8 inch record cards, ordered by grid coordinates. Even though the situation was very different to our current situation, the problem is the same: the metadata (grid coordinates) were not good enough for information retrieval, as users want a way to cross reference or search through the data (text on cards). ▇▇▇▇▇▇▇▇ sent out surveys by post asking archaeology professionals on their opinion on the use of computers for record manipulation, and found that 63% already did, or were hoping to do so in the future, meaning 37% of respondents did not see any value in using computers for this task. Eventually they concluded that “A computer-based recording system gives the potential to relieve problems of lack of space, lost data, inaccuracies in recording and to provide a flexible and efficient retrieval system, therefore relieving staff time for other work” (▇▇▇▇▇▇▇▇, 1983, p. 43), which is basically also the main aim of this project. It seems not much