Demographic Transition Model

The demographic transition model seeks to explain the transformation of countries from having high birth and death rates to low birth and death rates. In developed countries, this transition began in the eighteenth century and continues today. Less developed countries began the transition later and are still in the midst of earlier stages of the model. CBR CDR The model is based on the change in crude  birth rate (CBR) and crude death rate (CDR) over time. Each is expressed per thousand population. The CBR is determined by taking the number of births in one year in a country, dividing it by the countrys population, and multiplying the number by 1000. In 1998, the CBR in the United States is 14 per 1000 (14 births per 1000 people) while in Kenya it is 32 per 1000. The crude death rate is similarly determined. The number of deaths in one year is divided by the population and that figure is multiplied by 1000. This yields a CDR of 9 in the U.S. and 14 in Kenya. Stage I Prior to the Industrial Revolution, countries in Western Europe had high CBR and CDR. Births were high because more children meant more workers on the farm and with the high death rate, families needed more children to ensure the  survival of the family. Death rates were high due to disease and a lack of hygiene. The high CBR and CDR were somewhat stable and meant the  slow growth of a population. Occasional epidemics would dramatically increase the CDR for a few years (represented by the waves in Stage I of the model. Stage II In the mid-18th century, the death rate in Western European countries dropped due to improvement in sanitation and medicine. Out of tradition and practice, the birth rate remained high. This dropping death rate but the  stable birth rate at the beginning of Stage II contributed to skyrocketing population growth rates. Over time, children became an added expense and were less able to contribute to the wealth of a family. For this reason, along with advances in birth control, the CBR was reduced through the 20th century in developed countries. Populations still grew rapidly but this growth began to slow down. Many less developed countries are currently in Stage II of the model. For example, Kenyas high CBR of 32 per 1000 but low CDR of 14 per 1000 contribute to a high rate of growth (as in mid-Stage II). Stage III In the late 20th century, the CBR and CDR in developed countries both leveled off at a low rate. In some cases, the CBR is slightly higher than the CDR (as in the U.S. 14 versus 9) while in other countries the CBR is less than the CDR (as in Germany, 9 versus 11). (You can obtain current CBR and CDR data for all countries through the Census Bureaus International Data Base). Immigration from less developed countries now accounts for much of the population growth in developed countries that are in Stage III of the transition. Countries like China, South Korea, Singapore, and Cuba are rapidly approaching Stage III. The Model As with all models, the demographic transition model has its problems. The model does not provide guidelines as to how long it takes a country to get from Stage I to III. Western European countries took centuries through some rapidly developing countries like the Economic Tigers are transforming in mere decades. The model also does not predict that all countries will reach Stage III and have stable low birth and death rates. There are factors such as religion that keep some countries birth rate from dropping. Though this version of the demographic transition is composed of three stages, youll find similar models in texts as well as ones that include four or even five stages. The shape of the graph is consistent but the divisions in time are the only modification. An understanding of this model, in any of its forms, will help you to better understand population policies and changes in developed and less developed countries around the world.

What Foreign Aid Is Essay - 1730 Words

What Foreign Aid Is There are two words that many politicians like to shy away from, and those two words are, foreign aid. Taking a firm stand on either side of this topic is usually side stepped by decision makers. Their opinions are usually based on a case by case analysis. This extremely controversial topic involves whether or not to support the policy of foreign aid to needy or sometimes not so needy countries. What benefits does foreign aid have for the countries that receive it, and does it have any benefits for the countries who give? Some may say that instead of spending money on foreign aid, money should be spent on domestic aid. Those who argue in favor of foreign aid say that it is an investment in the future of†¦show more content†¦Some people would still be against this type of aid, and their argument would be that if money that goes to Mexico was invested in helping small business get started, America would be less dependent on other countries for their goods or services. In this case the nay-sayers were proven wrong because Mexico recently repaid the United States in full, plus interest, and a year in advance. Another form of foreign aid is a certain amount of credit is given to the receiving country in order to buy American made goods only. This form of aid not only helps the country in need, but it also pumps money into the US economy. Still some may wonder why should we care about a country such as Russia, to whom the U.S. has recently given aid to recently. The U.S. cant afford to have an ex-superpower unstable economically because of their nuclear capabilities. Because Russia is in the unstable beginnings of a democracy the U.S. wants to keep Russia on the right track. If that includes giving them aid in order to help feed their people, then that is what the U.S must do. Because reinstatement of the old regime would result in turmoil throughout the world. After communism was overthrown in the former Soviet Union the world went through big changes. There was no longer two world super powers, and all of a sudden a 40 year cold war ended. As stated the U.S. uses foreign aid as a tool to implement their political views in other countries.Show MoreRelatedWhat Role Does Foreign Aid Play? Developing Countries?1466 Words   |  6 PagesWhat Role Does Foreign Aid Play in Developing Countries? Many countries around the world receive foreign aid from the United States and other developed countries every year in large sums. This money is dispersed throughout the receiving country, it is the hope of the United States that this will help to create development economically and politically as well as promote the belief of democracy in these countries. The aid is intended to enhance infrastructure within the country so local institutionsRead MoreWhat Role Does Foreign Aid Play? Developing Countries?1768 Words   |  8 PagesWhat Role Does Foreign Aid Play in Developing Countries? Many countries around the world receive foreign aid from the United States and other developed countries every year in large sums. This money is dispersed throughout the receiving country, it is the hope of the United States that this will help to create development economically and politically as well as promote the belief of democracy in these countries. The aid is intended to enhance infrastructure within the country so local institutionsRead MoreForeign Countries Should Not Help Out A Less Developed Country?1348 Words   |  6 Pagesstudies have shown that giving money alone does not help, it could actually make the country worse off. Foreign countries should not give money to developing nations because the developing countries become too reliant on the developed countries, it does not help the developing countries, and with money, comes corruption. The better way for foreign aid to work which is to give technological aid. A quote from Maimonides, â€Å"Give a man a fish and y ou feed him for a day; teach a man to fish and you feedRead MoreThe United States Should Stop Sending Aid For Foreign Countries Essay1291 Words   |  6 Pageshave different views, but they both had the same opinion that the United States should stop sending aid to foreign countries. In this essay my view is that the United States should stop sending aid to foreign countries. I will defend my view against poverty by discussing Garret Hardin and, James Shikwati views on poverty, and how the United States and other countries will benefit from not sending aid. Garret Hardin mentions environmentalists use the metaphor of Earth as a spaceship. People persuadeRead MoreThe World Enters A Globalized Age Essay1151 Words   |  5 Pages As the world enters a globalized age, the interactions between nations become increasingly important. What was once a ‘all for one, one for all’ mentality, with nations acting mostly in their own best interests -barring, of course, the alliances made between nations, often military in nature acting against a common foe- has become a question of the common good. What obligation to help, if any, do countries with well developed economies and militaries have towards those nations in need? Nations thatRead MoreThe Effects of War and Peace on Foreign Aid in Somalia1360 Words   |  6 Pagesof War and Peace on Foreign Aid in Somalia Rachel Gardner Professor Marco Mena Sociology of Developing Countries Strayer University 5/4/2014 The Effects of War and Peace on Foreign Aid Foreign aid plays a major role in the sustainability of economic and social activities of developing nations. Whether in the form of development or humanitarian aid, the foreign aid is key to ensuring better living conditions and economic development for these states. These forms of aid are influential in pushingRead MoreForeign Aid Essay1342 Words   |  6 PagesForeign aid is defined as the help given by one nations government to another nation to help with social and financial advancement. Also, foreign aid is used in order to help a country to respond a disaster such as natural disaster, disease, etc. It can include providing financial help, medicines (science), food, clothes, etc. Likewise, foreign aid can be used for education (knowledge), infrastructures, equipment, and to fight poverty. I believe that foreign aid helps poor countries especiallyRead MoreThe Government Should Not Be Taxing Us For Someone Else s Needs1709 Words   |  7 Pagesto get rid of foreign aid. This paper is about how the government is taking our tax money and giving it to people in lower income countries. Us government should not be taxing us for someone else s needs because 1. People are already having trouble with taxes, and now that we added this it is going to be almost impossible for some people to live in this economy. 2. People that care about this should be funding this, not the people that have to pay taxes. 3. When we help them what do they assistRead MoreThe Beginning Of Foreign Aid Essay1505 Words   |  7 PagesIntroduction: 1.The beginning of Foreign Aid All human beings regardless of their background require a set of resources to survive. These are food, water, shelter and clothing in addition to this, healthcare, education and sanitation are also essential for a person’s wellbeing. In 1949 Truman’s Inaugural Address â€Å"we must embark on a bold new program for making the benefits of our scientific advances and industrial progress available for the improvement and growth of underdeveloped areas† raisedRead MoreCase Study : Bang For Their Buck Essay986 Words   |  4 PagesThis leads to ineffective aid, wasted time and money, and a fundamental sense of distrust and disenchantment with the system of developmental assistance. Easterly further argues that our approach to foreign aid has been that of a â€Å"Planner† instead of a â€Å"Searcher.† Planners are full of good intentions, but they very rarely motivate others; they raise expectations, but shirk responsibility for meeting them; most importantly, they apply â€Å"global blueprints† to foreign aid that universally aim to solve

A Poetry Analysis Justice Essay Example For Students

A Poetry Analysis Justice Essay In the piece Justice, the author included victims of notorious crimes in the Philippines who encountered at least any f the following: kidnap, rape, hazing, manslaughter, massacre, and murder. The literary piece Justice by Christine Caramel Villain is a poem written in the form of concrete poetry known as Carmen figurate, in which the words are typographically arranged in such a manner that a visual image is formed. In using shaped verse, the poet creates an illustration of a lifeline. The rises and falls, or the peaks (to represent the highs) and valleys (to represent the lows) of the fluctuation of the lifeline, which in electrocardiography (a noninvasive medical procedure for recording electrical activities of the heart) indicates that a certain individual is still living as characterized by his/her active heartbeat, are composed of names of victims murdered in the Philippines. The linear portion of the lifeline after a series of waves, however, is made up of iterations of the word Justice, which is written in a repetitive way for emphasis and further retention to the memory of the readers. In electrocardiography, the flat line implies that the beating of the heart has already terminated, thereby causing death. According to the author, the reason for the placement of the names within the yeoman part of the lifeline is to depict that the cases pertaining to the concerned people are yet unsolved and unclosed, or still withheld and hanging. As a result, the memory of the demise of the injured party lives on in spite of them being already deceased. On the other hand, the author has committed some minor mistakes in writing the names of the victims by way of misspelling a few of them (I. . , Chris instead of Crisp in Crisp Mended, Stateliest instead of Easterlies in Easterlies Eviction, Rockwell instead of Rachel in Rachel Groaned, and Regis instead of Regales in Marvin Regales). Other name entries are also mistaken for the reason that in one case, the surname linked to the first name is incorrect (Chapman instead of Hellman in Maureen Hellman) and in another, the first and s econd names are interchanged Cohn Roland instead of Roland John in Roland John Chapman). As for the organization of the names, the following sequence is maintained: Marjory Aching, Jacqueline Aching, Maureen Hellman, Roland John Chapman, Jesus Lenin, Nadia Blanch, Crisp Mended, Easterlies Eviction, Carmela Eviction, Jennifer Eviction, Geraldine Palm, Given Grace Sibilance, Ray Bernard Appearance, Rachel Groaned, Marvin Regales, and Marc Andre Marcos. However, the author failed to be consistent with the order of the names at the endmost part of the lifeline in which the last input is Marvin Regales instead of Rachel Groaned, provided that the order of the name list is the basis for succession. Other errors marked in the work include the failure to insert Roland John Chapman in between Maureen Hellman and Jesus Lenin at the first sequence of the names. Other names are as well deficient such as Biz at the second sequence, Men at the third sequence, and Erg at the latter sequence mentioned. Space fillers are also added so as not to ruin the figure and to discipline the use of names t fit the form (such as ma after Groaned at the second sequence and sad after Given at the third sequence). But this does not actually suffice and is not proper since the text should be as important as the shape, hence being called Carmen figurate (visual poem) and not merely Carmen (poem) or figurate (image) alone. If fillers are supplied, the main purpose of the poetry, which is to preserve congruity between the figure and the content, is by all means defied. As for the spatiality of the poem, however, commas or spaces are not used to separate one person from another. The names are keyed in one after another and without interruption. Tell all the Truth but Tell it Slant by Emily Dickinson analysis EssayEven the seven-time repetition of the word Justice at the flat segment of the lifeline is written with no pauses at all. Punctuation marks are not even utilized per SE. On the other hand, although names are written repetitively to signify importance, the poem might have been more effective if it is appropriately punctuated with the aid of commas for short and slight pauses, or semicolons for the intention of a caesura or an extended pause before shifting from a thought or idea t the succeeding one. However, if the aim is to establish a complete stop, a period can be employed. Also, it would be of assistance if the initial letter of the names is write in capital letters to show that each person is important. Moreover, the upper and lower ends of the zigzag lines should entirely carry the names of the victims instead of breaking then placing them in separate lines. Aside from the order of the names, however, there is little attempt at harmonious arrangement (Holcomb, 2007) f the poetry. There is no other particular pattern observed than the mere arrangement of the names. The author, in this literary work, used run-on lines that are suggestive of continuous reading. Lastly, the objective of the poetry, which is to find meaning and explanation yet not in the literal sense is accomplished, since the concept of Justice is embedded within an illustration of a lifeline. This suggests, as represented in the graph, that Justice in the Philippines is dead, as the cases of the abused involved in murder have still not progressed over the years. This may also mean that criminal Justice in the country is already limping, for some witnesses of the crimes are unwilling to lodge complaints, support prosecutions, (Asian Human Rights Commission).

Web Portal Essay Example

Web Portal Essay Automatic Identification of Temporal Information in Tourism Web Pages Stephanie Weiser*, Philippe Laublet**, Jean-Luc Minel* * MoDyCo, UMR 7114, CNRS 200 avenue de la Republique, 92001 Nanterre ** LaLIC, Universite Paris-Sorbonne Maison de la recherche, 28 rue Serpente 75006 Paris E-mail: steph. [emailprotected] com, Philippe. [emailprotected] fr, [emailprotected] fr Abstract This paper presents our work on the detection of temporal information in web pages. The pages examined within the scope of this study were taken from the tourism sector and the temporal information in question is thus particular to this area.The differences that exist between extraction from plain textual data and extraction from the web are brought to light. These differences mainly concern the spatial arrangement of the text, the use of punctuation and the respect of traditional syntactic rules. The temporal expressions to be extracted are classified into two kinds: temporal information that concerns one parti cular event and repetitive temporal information. We adopt a symbolic approach relying on patterns and rules for the detection, extraction and annotation of temporal expressions; our method is based on the use of transducers.First evaluations have shown promising results. Since the visual structure of a web page is very important and often informs the user before he has even read the text, a semiotic study is also presented in this paper. 1. Introduction With the methods of the Semantic Web, portal applications can be created, relying on ontologies. For these applications and many service applications, temporal information is often essential. For example, a tourism web portal would need information about the type of tourism object and its location in time and space.In addition, the extracted information must be stored in the knowledge base according to the ontology used by the application. In this paper we will focus on temporal information in tourism web pages. The temporal informat ion has to be detected, extracted and annotated. The annotation format will probably rely on existing XML tools (Stern 2007). To perform these tasks, we encountered three main kinds of difficulties. First, we have to deal with complex, imprecise temporal information. Of course, single dates are easy to process but more complex expressions, such as periods or repetitive information (e. . from March to July, open every day except Tuesday), must be treated as well. Second, after being extracted, the information needs to be linked to the proper tourism object. If the web page concerns only one object, this is straightforward, but some web pages concern many objects and an analysis is therefore necessary to decide how to link each piece of information to its object. Third, the web pages we deal with are all of the same type: tourism web pages. However, they vary a lot as they are made by different people, have different forms and concern different types of tourism objects.We will try to show that a semiotic study of some pages is necessary to take into account some of their specificities. The work presented in this paper is situated within the framework of the EIFFEL project. Its main objective is to create a portal in the area of tourism with different functionalities. This portal, for use by the local tourism sector, will include a specialised search engine. It will allow users to find and collect precise and essential information in context. It will also help the territory as a French region to promote its services. This is a wideranging project (Noel et al. 008) which is based on web semantic technologies, knowledge representation and linguistic methods and expertise. It includes automatic identification, selection and extraction of various items from the Web according to existing ontologies. This project involves mainly two companies – Mondeca and Antidot – and three laboratories – LIRMM (CNRS), INRIA-Rocquencourt and MoDyCo (CNRS). Our co rpus, meaning that the pages have been collected for this particular study, in a precise context, is composed of more than 5000 tourism web pages in French. These pages were collected by automatic crawling of websites.They were collected by the company Antidot and transformed into XML files in order to make them more suitable for automatic processing. 2. Temporal Information in Web Pages The detection and extraction of temporal information is a subject that has already been widely studied (Battistelli et al. 2006), but mostly in the context of â€Å"printed textual objects† and not in texts specifically written for web pages. One of the aims of this paper is to show that studying the same type of temporal information is very different when it takes place in a web context.In tourist guide books, for example, the information is a lot more structured and standardized than on our web sites, which are extremely diverse and do not follow a 127 common page model. Every entry of a gu ide concerning a particular tourism object is often structured in the same way and contains the same type of information. Linguistic markers allow us to locate each piece of information. Punctuation is very rigorous and facilitates interpretation, especially automatic interpretation. On the other hand, on web pages, punctuation – if there is any – is rarely correctly used.More often, white spaces or new lines are used as separators. Even if the analysis of a text is not easy, a lot of syntactic tools already exist to perform this task automatically. For example, it is possible to build the syntactic tree of a given sentence, find the verb, etc. However, some complicated tasks, such as anaphora resolution, can make the analysis of texts a lot more complex. What we are interested in here is that the difficulties that are encountered for web pages are very different from those that text analysis presents.For example, in a tourism guide book1, each restaurant entry is stru ctured in the same way: name, address, phone number, metro station and opening details, e. g. ouvert tous les jours (open every day). Each item can easily be automatically processed. The problem is that, on a web page, the same kind of information would more likely look like the web page presented in figure 1. integrating temporal and spatial reasoning into XML query and transformation operations. Their work has lead to WebCal, a program which provides calendrical calculations for web services. 3. Tool and Method for the Temporal Expression DetectionWe have a symbolic approach relying on patterns and rules for the detection, extraction and annotation of temporal expressions (in French). We do not use any machine learning techniques. Our method relies on the use of transducers, created using the graphical interface of the Unitex corpus processor2. Unitex provides an intuitive and user-friendly method for creating transducers, which tag the output (as opposed to regular expressions wh ich do not modify the output). A wide variety of patterns can be recognized and tagged by our transducers, from very simple expressions to very complex ones.On one hand, generalizations can be made (such as using the symbol to tag all numbers). On the other hand, the transducers can be very detailed and precise for particular cases, for example, to recognize expressions such as a lexception du mois de mai – except in May. Here are a few examples of the expressions to be detected: (1) Le 21 avril April, the 21st (2) Du 10 au 21 mars From the 10th to the 21st of March (3) Du lundi au vendredi, 9h – 11h From Monday to Friday, 9am-11am (4) Inauguration du musee le 7 juillet 2006 Inauguration of the museum on the 7th of July 2006 (5) En Juillet et Aout ouvert tous les jours.Hors cette periode, fermeture le mercredi soir, le jeudi toute la journee et le dimanche soir. Horaires douverture: de 12h00 a 14h00 de 19h00 a 22h00. Open every day in July and August. Outside this peri od, closed on Wednesday night, all day Thursday and on Sunday night. Opening times: from noon to 2pm from 7pm to 10pm. Figure 1: Example For now, there is no tool able to interpret that Chez Paul et Bernadette Colas is the name of a Bed Breakfast or that Ouvert du 15 fevrier au 30 (open from February 15th to 30th) is its opening information.This is why the extraction of temporal information from texts can not be directly applied for Web pages: there is no standardized â€Å"syntax† for web pages. Our work is close to that of (Tenier et al. 2006). The object of their work is to extract, from web pages, information about contributors to scientific events. Our approach is more general since the web pages analyzed by these authors follow spatial arrangement rules while spatial arrangement on our pages is completely free. Our work is also close to that of (Bry et al. 2003) who aim at 1 3. 1. ClassificationAs shown in these examples, the type of temporal information to be identifi ed varies. Two main types can be highlighted: temporal information that concerns one particular event and repetitive temporal information. For the first type there are dates (concert on October 1st), periods (festival from May to June), and times (the concert starts at 8:00). For the second there are times (the museum opens at 10am), periods (the restaurant is open from Monday to Saturday) and exceptions (the camping ground is open all year long except in January). More complex examples, such as from MarchLe guide du routard – Paris balades 2007 p. 231 :  « Le SaintAmour 2, av. Gambetta, 75020. 01-47-97-20-15. Metro PereLachaise. Ouvert tous les jours  » (Open every day) 2 Unitex: http://www-igm. univ-mlv. fr/~unitex 128 to July, open every day except Tuesday, can also be included in this classification. In addition, not every occurrence of temporal information needs to be detected, since they do not necessarily concern a tourism object. For example, a single date, which is not introduced by a linguistic marker such as heure douverture (opening time) probably does not concern a tourism object.Single dates like these often concern the web page itself as in Derniere modification : 25/10/2006 (last modification : 25/10/2006). Therefore, we have been careful not to detect these dates. One problem that has been encountered is that some lexical marks are ambiguous. For example is Friday and Sunday night to be interpreted as meaning Friday night and Saturday night, or rather Friday all day and Saturday night? The context is therefore necessary to remove the ambiguity. For this example, if the object is a concert, the first interpretation will probably be the right one.For the whole corpus of 100 pages, we obtain a recall rate of 92% and a precision of 76%. Figure 2: Recall for the first evaluation 3. 2. Technical Information The tool Unitex, which uses dictionaries based on the tables of the LADL3, allows the user to process corpora on the lexical, syntact ic and morphological levels. It allows for the identification and tagging of structures which correspond to regular expressions, represented by finite state graphs. This graphical representation makes using Unitex more intuitive than the use of long regular expressions.We built our transducers using this tool, following the linguistic study that listed the patterns of temporal expressions that could be found on web pages. The set of transducers includes one main transducer and 23 secondary graphs. These transducers include 88 lexical markers (like ouverture – opening, mars – March, inauguration) and 22 grammatical markers (like sur – on, le – the, dans – in). Figure 3: Precision for the first evaluation The important variation in the results is due to the fact that there are not a lot of temporal expressions to detect in the pages (7 for sets 1 and 3, 9 for set 2, 20 for set 4 and 8 for set 5).The precision for set 5 is especially low because one p age contains a forum with 4 temporal expressions that are detected but do not concern a tourism object. 4. Evaluations 4. 1. First Evaluation The transducers allowing the detection of temporal information were manually created from a study of a set of pages. In order to evaluate and optimise them, we tested them on sets of 20 web pages, randomly manually selected from our corpus. This operation was performed five times and the transducers have thus been tested on 100 different pages. The results of this experiment are presented in the diagrams below.Recall and precision were calculated for each set of 20 pages (figure 2 and figure 3). 4. 2. Second Evaluation A different procedure was used for the second evaluation. We deliberately manually selected 25 pages containing temporal information. Some of these pages were randomly opened and selected when they were found to contain temporal information; for the others our transducers were applied to more than 200 pages and those with result s were selected. For these 25 pages, the recall is 39,1% and the precision is 46,5%. Here are a few numbers to explain these figures. These pages contain 69 expressions to detect.The transducers match and tag 58 expressions: 27 are correctly detected, 3 are falsely detected (they should not be detected because they do not correspond to temporal tourism information), 28 expressions are incompletely detected and 25 expressions are missed. We are faced with two kinds of incomplete expressions. In the first kind a part of the expression is not matched, Laboratoire d’Automatique Documentaire et Linguistique – the tables were created by Maurice Gross and consist in a classification of lexical units according to syntactic and distributional criteria. 3 129 or example in toute la journee de 8h00 a 19h00 (all day long from 8:00am to 7:00pm), only the underlined part is detected. In the second kind of incomplete expressions, each pattern to be matched is detected in two or more matches, for example ouvert tous les jours en juillet et aout (open every day in July and August) is detected in two matches. 16 of the 28 incomplete expressions produced by our system correspond in fact to 5 correct expressions. Horaires Lundi Mardi Mercredi Jeudi Vendredi Samedi Matin 8h30-12h00 8h30-12h00 8h30-12h00 8h30-12h00 8h30-12h00 10h00-11h30Ap-midi 14h00-18h00 14h00-18h00 14h00-18h00 14h00-18h00 14h00- Figure 5: vertical table of opening times A human user would understand both of these tables without the slightest problem. For an automatic treatment, however, it is more complicated. If we do not look at the structure, but only at the text, the first table would then contain â€Å"Horaires lundi mardi mercredi jeudi vendredi samedi matin 08h30 12h00 08h30 12h00 08h30 12h00 08h30 12h00 08h30 12h00 10h00 11h30 ap. midi 14h00 18h00 14h00 18h00 14h00 18h00 14h00 18h00 14h00 †: a text flow that is incomprehensible.The second table would contain â€Å"Horaires Matin Ap-midi Lundi 8h30-12h00 14h0018h00 Mardi 8h30-12h00 14h00-18h00 Mercredi 8h3012h00 14h00-18h00 Jeudi 8h30-12h00 14h00-18h00 Vendredi 8h30-12h00 14h00 – Samedi 10h00-11h30 † it is clearer but it is very different from the text of the first one, for exactly the same meaning. Technical tools that allow a specific markup for headers in the table exist. But the web pages of our corpus are mostly built by professionals from the tourism sector and not by professional web designers: they do not use these technical tools.Consequently, only graphical elements and lexical content allow interpretation of these kinds of tables. In order to take into account these semiotic marks in automatic processing, the structure of the pages must be studied. An analysis of the XML tree is therefore necessary (Tenier et al. 2006). The high number of missed expressions is due to the fact that isolated dates without context are voluntarily ignored in order to avoid incorrect detections (li ke in derniere modification : 25 avril 2006 – last modification: April the 25th 2006).Some pages presenting an event schedule contain more than 10 dates without context that are therefore not detected. 22 out of the 25 missed expressions are concerned by this problem, which will be solved in the near future. 5. The Semiotic Marks For now, only textual information is taken into account, but a web page presents information in many other ways. The visual structure of a web page often informs the user before he has even read the text. For example, the structure could allow him to understand that the page concern one single object and not a collection of objects.In the same way, the formatting style (font, size, colour) can be very meaningful since it can often be assumed that two items written in the same font and colour are linked. One difficulty lies in automatically interpreting these semiotic marks. Many things are understandable at first sight for a human being but are hard to specify for automatic processing. On web pages, tables are commonly used to present information. They need a special treatment since their interpretation is not straightforward. 6. Conclusion In this paper we have presented our work on the detection of temporal information in web pages.We have focused on the differences that exist between extraction from plain textual data and extraction from the web. In addition, we introduced our main linguistic resource, our transducers, which could be used for many different NLP applications. We also presented the results obtained in applying these transducers to our corpus. The transducers that have been presented will be improved. They will be enriched with information concerning the type of tourism object, in order to determine the links between the temporal information detected and what they refer to.The work on semiotic marks will be continued in order to implement the study of the structure of the pages, analysing XML trees. Figure 4: O pening times table The figure 4 is an example of a table of opening times found on a web page. The first row contains the names of the days, the second row contains the opening times in the mornings and the last row contains the opening times in the afternoons. The information could have been presented in a vertical table as the one shown in figure 5: 130 7. Acknowledgements The project EIFFEL (ANR-05-RNTL-007) is funded with an ANR grant. .References Battistelli, D. , Minel, J. -L. , Schwer, S. (2006). Representation des expressions calendaires dans les textes : une application a la lecture assistee de biographies, Traitement Automatique des Langues, 47, 3, pp. 126. Bry, F. Lorenz, B. Ohlbach, H. J. Spranger, S. (2003). On Reasoning on Time and Location on the Web, Lecture Notes in Computer Science, Springer-Verlag, Germany, pp. 6983. Noel, L. , Carloni, O. , Moreau, N. , Weiser, S. (2008). Designing a Knowledge-Based Tourism Information System, Int. J. f Digital Culture and Electr onic Tourism, Special Issue on National Tourism Organisations and Exploitation of Information Technologies, to be published. Stern, R. -D. (2007). Expression linguistique du temps et representation ontologique : OWL-Time et etude des adverbiaux temporels, Memoire de Master IILGI, Universite de Paris-Sorbonne. Tenier, S. , Toussaint, Y. , Napoli, A. et Polanco, X. (2006). Instantiation of relations for semantic annotation, In the 2006 IEEE/WIC/ACM International Conference on Web Intelligence WI 2006, pp. 463-472 131