Higher Quality Input Phrase To Driven Reverse Dictionary

Higher Quality Input Phrase To Driven Reverse Dictionary Implementing a Higher Quality Input Phrase To Driven Reverse Dictionary E.Kamalanathan  and C.Sunitha Ram ABSTRACT Implementing a higher quality input phrase to driven reverse wordbook. In contrast to a conventional forward wordbook, that map from word to their definitions, a reverse wordbook takes a user input phrase describing the specified construct, and returns a group of candidate words that satisfy the input phrase. This work has important application not just for the final public, notably those that work closely with words, however conjointly within the general field of abstract search. The current a group of algorithms and therefore the results of a group of experiments showing the retrieval accuracy and therefore the runtime latency performance is implementation. The experimental results show that, approach will offer important enhancements in performance scale while not sacrificing the standard of the result. Experiments scrutiny the standard of approach to it of presently on the market reverse dictionaries show that the approach will offer considerably higher quality over either of the opposite presently on the market implementations. Index Terms : Dictionaries, thesauruses, search process, web-based services. . INTRODUCTION A Report work on creating a reverse dictionary, As against a regular (forward) wordbook that maps words to their definitions, a WD performs the converse mapping, i.e., given a phrase describing the required conception, it provides words whose definitions match the entered definition phrase. It’s relevant to language understanding. The approach has a number of the characteristics expected from a strong language understanding system. Firstly, learning solely depends on unannoted text information, which is abundant and contain the individual bias of an observer. Secondly, the approach is predicated on all-purpose resources (Brill’s PoS Tagger, WordNet [7]), and also the performance is studied below negative (hence additional realistic) assumptions, e.g., that the tagger is trained on a regular dataset with doubtless totally different properties from the documents to be clustered. Similarly, the approach studies the potential advantages of victimization all potential senses (and hypernyms) from WordNet, in an endeavor to defer (or avoid altogether) the necessity for Word Sense Disambiguation (WSD), and also the connected pitfalls of a WSD tool which can be biased towards a particular domain or language vogue BACKGROUND WORK Natural Language Processing: Natural Language Processing (NLP) [6] is a large field which encompasses a lot of categories that are related to this thesis. Specifically NLP is the process of computationally extracting meaningful information of natural languages. In other words: the ability for a computer to interpret the expressive power of natural language. Subcategories of NLP which are relevant for this thesis are presented below. WordNet: WordNet [7], [2]is a large lexical database containing the words of the English language. It resembles the traits of a thesaurus in that it structures words that have similar meaning together. WordNet is something more, since it also specifies different connections for each of the senses of a given word. These connections place words that are semantically related close to one another in a network. WordNet also displays some quality of a dictionary, since it describes the definition of words and their corresponding part-of-speech. Synonym relation is the main connection between words, which means that words which are conceptually equivalent, and thus interchangeable in most contexts, are grouped together. These groupings are called synsets and consist of a definition and relations to other synsets. A word can be part of more than one synset, since it can bear more than one meaning. WordNet has a total of 117 000 synsets, which are linked together. Not all synsets have a distinct path to another synset. This is the case, since the data structure in WordNet is split into four different groups; nouns, verbs, adjectives and adverbs (since they follow different rules of grammar). Thus it is not possible to compare words in different groups, unless all groups are linked together with a common entity. There are some exceptions which links synsets cross part-of-speech in WordNet, but these are rare. It is not always possible to find a relation between two words within a group, since each group are made of different ba se types. The relations that connect the synsets within the different groups vary based on the type of the synsets. Application Programming Interface Several Application Programming Interfaces (API) exists for WordNet. These allow easy access to the platform and often additional functionality. As an example of this the Java WordNet Library [8] (JWNL) can be mentioned. This allows for access to the WordNet Library files. PoS Tagging PoS tags[8] are assigned to the corpus using Brill’s PoS tagger. As PoS tagging require the words to be in their original order this is done before any other modifications on the corpora. Part-of-speech (POS) tagging is the field which is concerned with analysing a text and assigning different grammatical roles to each entity. These roles are based on the definition of the particular word and the context in which it is written. Words that are in close proximity of each other often affect and assign meaning to each other. The POS taggers job is to assign grammatical roles such as nouns, verbs, adjectives, adverbs, etc. based upon these relations. The tagging of POS is important in information retrieval in general text processing. This is the case since natural languages contain a lot of ambiguity, which can make distinguishing words/terms difficult. There are two main schools when tagging POS. These are rule-based and stochastic. Examples of the two are Brill’s tagger and Stanford POS tagger, respectively. Rule-based taggers work by applying the most used POS for a given word. Predefined/lexical rules are then applied to the structure for error analysis. Errors are corrected until a satisfying threshold is reached. Stochastic taggers use a trained corpus to determine the POS of a given word. Stopword Removal Stopwords, i.e. words thought not to convey any meaning, are removed from the text. The approach taken in this work does not compile a static list of stopwords, as usually done. Instead PoS information is browbeaten and all tokens that are not nouns, verbs or adjectives are removed. Stop words are words which occur often in text and speech. They do not tell much about the content they are wrapped in, but helps humans understand and interpret the residue of the content. These terms are so generic that they do not mean anything by themselves. In the context of text processing they are basically just empty words, which only takes up space, increases computational time and affects the similarity measure in a way which is not relevant. This can result in false positives. Table: 1 List of Stop words This class includes only one method; which runs through a list of words and removes all occurrences of words specified in a file. A text file, which specifies the stop words, is loaded into the program. This file is called â€Å"stop-words.txt† and is located at the home directory of the program. The text file can be edited such that it only contains the desired stop words. A representation of the stop words used in the text file can be found in table 1. After the list of stop words has been loaded, it is compared to the words in the given list. If a match is found the given word in the list is removed. A list, exposed from stop words, is then returned. Stemming Words with the same meaning appear in various morphological forms. To capture their similarity they are normalised into a common root-form, the stem. The morphology function provided with WordNet is used for stemming, because it only yields stems that are contained in the WordNet dictionary. This class contains five methods; one for converting a list of words into a string, two for stemming a list of words and two for handling the access to WordNet through the JWNL API[8]. The first method listToString() takes an ArrayList of strings and concatenate these into a string representation. The second method stringStemmer() takes an ArrayList of strings and iterates through each word, stemming these by calling the private method wordStemmer(). This method checks if the JWNL API has been loaded and starts stemming by looking up the lemma of a word in WordNet. Before this is done, each word starting with an uppercase letter is checked to see if it can be used as a noun. If the word can be used as a noun, it does not qualify for stemming and is returned in its original form. The lemma lookup is done by using a morphological processor, which is provided by WordNet. This morphs the word into its lemma, after which the word is checked for a match in the database of WordNet. This is done by running through all the specified POS databases defined in WordNet. If a match is found, the lemma of the word is returned, otherwise the original word is simply returned. Lastly, the methods allowing access to WordNet initializes the JWNL API and shuts it down, respectively. The initializer() method gets an instance of the dictionary files and loads the morphological processor. If this method is not called, the program is not able to access the WordNet files. The method close() closes the dictionary files and shuts down the JWNL API. This method is not used in the program, since it would not make sense to uninstall the dictionary once it has been installed. It would only increase the total execution time. It has been implemented for good measure, should it be needed. Stemming[5] is the process of reducing an inflected or derived word to its base form. In other words all morphological deviations of a word are reduced to the same form, which makes comparison easier. The stemmed word is not necessarily returned to its morphological root, but a mutual stem. The morphological deviations of a word have different suffixes, but in essence describe the same. These different variants can therefore be merged into a distinct representative form. Thus a comparison of stemmed words turns up a higher relation for equivalent words. In addition storing becomes more effective. Words like observes, observed, observation, observationally should all be reduced to a mutual stem such as observe. PROPOSED SYSTEM Reverse dictionaries approach can provide significantly higher quality. The proposed a set of methods for building and querying a reverse dictionary. Reverse dictionary system is based on the notion that a phrase that conceptually describes a word should resemble the word’s actual definition, if not matching the exact words, then at least conceptually similar. Consider, for example, the following concept phrase: â€Å"talks a lot, but without much substance.† Based on such a phrase, a reverse dictionary should return words such as â€Å"gabby,† â€Å"chatty,† and â€Å"garrulous.† Forward mapping (standard dictionary): Intuitively, a forward mapping designates all the senses for a particular word phrase. This is expressed in terms of a forward map set (FMS). The FMS of a (word) phrase W, designated by F(W) is the set of (sense) phrases {S1, S2, . . . Sn } such that for each Sj à Ã¢â‚¬Å¾ F(Wi), (Wi à ¯Ã†â€™Ã‚   Sj) à Ã¢â‚¬Å¾ D. For example, suppose that the term â€Å"jovial† is associated with various meanings, including â€Å"showing high-spirited merriment† and â€Å"pertaining† to the god Jove, or Jupiter.† Here, F (jovial) would contain both of these phrases. Reverse mapping (reverse dictionary): Reverse mapping applies to terms and is expressed as a reverse map set (RMS). The RMS of t, denoted R(t), is a set of phrases { P1, P2, Pi,†¦Ã¢â‚¬ ¦, Pm}, such that à ¯Ã¢â€š ¬Ã‚ ¢Pi à ¯Ã¢â€š ¬Ã‚  Ãƒ ¯Ã†â€™Ã… ½ R(t), t à ¯Ã†â€™Ã… ½ F(Pi). Intuitively, the reverse map set of a term t consists of all the (word) phrases in whose definition t appears. The find candidate words phase consists of two key sub steps: 1) Build the RMS. 2) Query the RMS. A. COMPONENTS The first preprocessing step is to PoS tag the corpus. The PoS tagger relies on the text structure and morphological differences to determine the appropriate part-of-speech. For this reason, if it is required, PoS tagging is the first step to be carried out. After this, stopword removal is performed, followed by stemming. This order is chosen to reduce the amount of words to be stemmed. The stemmed words are then looked up in WordNet and their corresponding synonyms and hypernyms are added to the bag-of-words. Once the document vectors are completed in this way, the frequency of each word across the corpus can be counted and every word occurring less often than the pre specified threshold is pruned. Stemming, stopword removal and pruning all aim to improve clustering quality by removing noise, i.e. meaningless data. They all lead to a reduction in the number of dimensions in the term-space. Weighting is concerned with the estimation of the importance of individual terms. All of these have been used extensively and are considered the baseline for comparison in this work. However, the two techniques under investigation both add data to the representation. a PoS tagging adds syntactic information and WordNet is used to add synonyms and hypernyms. B. BUILDING REVERSE MAPPING SETS The input phrases sentence is split into words and then removes the stop words ( a, be, person, some, someone, too, very, who, the, in, of, and, to) if any appears, and find other words, which is having same meaning from the forward dictionary data sources. Given the large size of dictionaries, creating such mappings on the fly is infeasible. Thus, Procreate these Rs for every relevant term in the dictionary. This is a one time, offline event; once these mappings exist, we can use them for ongoing lookup. Thus, the cost of creating the corpus has no effect on runtime performance. For an input dictionary D, we create R mappings for all terms appearing in the sense phrases (definitions) in D. C. RMS QUERY This module responds to user input phrases. Upon receiving such an input phrase, we query the R indexes already present in the database to find candidate words whose definitions have any similarity to the input phrase. Upon receiving an input phrase U, we process U using a stepwise refinement approach. We start off by extracting the core terms from U, and searching for the candidate words (Ws) whose definitions contain these core terms exactly. (Note that we tune these terms slightly to increase the probability of generating Ws) If this first step does not generate a sufficient number of output Ws, defined by a tuneable input parameter ÃŽ ±, which represents the minimum number of word phrases needed to halt processing and return output. D. CANDIDATE WORD RANKING In this module sorts a set of output Ws in order of decreasing similarity to U, based on the semantic similarity. To build such a ranking, we need to be able to assign a similarity measure for each (S,U) pair, where U is the user input phrase and S is a definition for some W in the candidate word set O. Wn and Palmer’s Conceptual similarity, WUP Similarity between concepts a and b in a hierarchy, Here depth(lso(a,b)) is the global depth of the lowest super ordinate of a and b and len(a,b) is the length of the path between the nodes a and b in the hierarchy SOLUTION ARCHITECTURE We now describe our implementation architecture, with particular attention to design for scalability. The Reverse Dictionary Application (RDA) is a software module that takes a user phrase (U) as input, and returns a set of conceptually related words as output. Figure 1. Architecture of reverse dictionary. The user input phrase, split the word from the input phrase, perform the stemming. Predict every relevant term in the forward dictionary data source. In the generate query. input phrase, minimum and maximum output thresholds as input, then removal of level 1 stop words ( a, be, person, some, someone, too, very, who, the, in, of, and, to) and perform stemming, generate the query.Execute the query find the set of candidate words. Finally sort the result based on the semantic similarity EXPERIMENTAL ENVIRONMENT Our experimental environment consisted of two 2.2 GHz dual-core CPU, 2 GB RAM servers running Windows XP pro and above. On one server, we installed our implementation our algorithms (written in Java). The other server housed is wordnet dictionary data. CONCLUSION We describe the many challenges inherent in building a reverse lexicon, and map drawback to the well-known abstract similarity problem. We tend to propose a collection of strategies for building and querying a reverse lexicon, and describe a collection of experiments that show the standard of our results, similarly because the runtime performance underneath load. Our experimental results show that our approach will give important enhancements in performance scale while not sacrificing answer quality. The higher quality input phrase to driven reverse dictionary. Unlike a traditional forward dictionary, which maps from words to their definitions, a reverse dictionary takes a user input phrase describing the desired concept, it reduce the well-known conceptual similarity problem. The set of methods building a reverse mapping querying a reverse dictionary and it produces the higher quality of results. This approach can provide significant improvements in performance scale without sacrificing solution quality but for larger query it is fairly slow. REFERENCES T. Dao and T. Simpson, â€Å"Measuring Similarity between Sentences,† 2009. http://opensvn.csie.org/WordNetDotNet/trunk/ Projects/ T. Hofmann, â€Å"Probabilistic Latent Semantic Indexing,† SIGIR ’99: Proc. 22nd Ann. Int’l ACM SIGIR Conf. Research and Development in Information Retrieval, pp. 50-57, 1999. D. Lin, â€Å"An Information-Theoretic Definition of Similarity,† Proc .Int’l Conf. Machine Learning, 1998. M. Porter, â€Å"The Porter Stemming Algorithm,†http://tartarus.org/martin/PorterStemmer/ , 2009. G. Miller, C. Fellbaum, R. Tengi, P. Wakefield, and H. Langone, â€Å"Wordnet Lexical Database,† http://wordnet.princeton.edu/wordnet/download/, 2009. P. Resnik, â€Å"Semantic Similarity in a Taxonomy: An Information-Based Measure and Its Application to Problems of Ambiguity in Natural Language,† J. Artificial Intelligence Research, vol. 11, pp. 95- 130, 1999. AUTHORS PROFILE E Kamalanathan is pursuing his Master of Engineering (part time ) from Department of Computer Science and Engineering, SCSVMV University Enathur,

The Politics of Name Changes in India Essay -- Town City

The Politics of Name Changes in India Ever since independence in 1947, many locations in India have changed their names. Much of this resulted from the reorganization of the states on linguistic lines (as opposed to British colonial divisions). However, in the last six years, many major towns and cities have been renamed in ways that affect foreigners more. Among this flood of changes, three stand out. These are the former cities of Bombay, Madras, and Calcutta, which, together with Delhi, are considered the "mega cities" of India.1 They are the four most populous cities in India, and all but Madras are among the 15 most populous cities in the world.2 As a result, they are important commercial and transit hubs, and are well known outside India. Yet nearly six years later, most non-Indians still have no idea that they are now named Mumbai, Chennai, and Kolkata. Given the difficulties involved in these changes, one expects compelling justifications for the changes. In each case, these changes have officially been justified on anti-colonialist grounds. However, I will argue that these changes are instead tools for channeling regionalist sentiment in the conflict between the Hindu nationalist Bharatiya Janata Party (BJP), the centrist Indian National Congress (INC), and various regional parties. Opponents of the INC proposed all these names, but those proposed by the BJP and its nationalist allies have been more divisive. The case of Mumbai was the first major change to happen, and is thus the best known. In 1995, the ruling party in the government of the state of Maharashtra (of which Bombay was capital) announced that Bombay's name would be changed to its Marathi name, Mumbai.3 The casual observer would assume that the name "B... ...ttp://www.hartford-hwp.com/archives/52a/076.html - a BBC article about the Kolkata name change * http://www.lonelyplanet.com/destinations/indian_subcontinent/mumbai/history.htm - a travel site with a history of Mumbai. (It connects Bal Thackeray, the leader of the Shiv Sena, to Adolf Hitler!) * http://theory.tifr.res.in/bombay/history/people/language/ - a list of most languages in India spoken by over a million people * http://www.contactsindia.com/tourism/south/tamil_nadu/madras.htm - a site with information about Chennai * www.aapkashahar.com - a tourist site with little or no graphics, demonstrating usage of both old and new names for cities * www.soulkurry.com - a site about women's spirituality in India List of Indian cities that have recently undergone (or may undergo) a name change. (Population figures are from www.citypopulation.de.)

Assess the View on Whether or Not Roles in Relationships

H/W Assess the view that roles and relations among couples are becoming more equal (24 Marks) Numerous sociologists have suggested that a large number of relationships are now becoming more symmetrical in compare to the traditional families looking back 40, 100, 200 years ago. They say that the traditional male and female roles are no longer as they were before, it has all fallen apart, and hence relationships have been becoming more equal. Some sociologists such as Wilmot and Young argue that in symmetrical families, normal domestic duties are most likely to be shared by both parties and they are both very likely to be working.Thus it all undoes the tradition of the past in which a woman would care for the children and the home, and the man being the ‘breadwinner’ would head out and work. Some sociologists also argue that the decision making is no longer only within the man’s hand as it once was, but also in the woman’s hands, the symmetrical family has an equal partnership in decision making. All this evidence leads some sociologists to view the roles and relations of couples to have been becoming somewhat more equal in compare to the past.On the other hand, some sociologists such as Morgan(1996) argue that women now take part in an act known as the ‘Triple Shift’, this is when women go out and work, but then also do domestic work when at home and give emotional support to partner and children. As a result, many feminists would argue that the roles and relations of couples are not equal, but actually unfair, the woman is doing so much more than her partner. This views result in some sociologist believing that roles and relations have changed in compare to the past, but in a path towards the woman doing more work than the man, making her the ‘breadwinner’.In conclusion, many sociologists believe that the roles and relations of couples have been becoming more equal, but there are some who still view it to not being equal at all, and some even arguing that women are doing more work than the men. From this evidence it is clear to say that it is not clear whether or not roles and relations are becoming more ‘equal’, but many sociologists argue that there has been a big change in compare to the past traditions, but whether or not it’s towards the path of the ‘more equal’ is still being argued by many. Siad Mohamed Siyad

Mate Selection Theories essays

Mate Selection Theories essays To describe men and women's differential preferences in mate selection, I am going to use the following theories: 1) sexual selection theory, 2) parental investment theory, and 3) social learning theory, and 4) sexual strategies theory. All of these theories have been used to explain why men and women have differential preferences in mate selections. In the evolutionary approach, biological fitness is defined as reproductive success. Therefore, understanding particular behavioral sex differences is of great interest to evolutionists. Reproduction is central to the evolutionary process and no domain is closer to reproduction than sexuality. Darwin (58) adopted phrase "survival of the fittest" to summarize the process of natural selection, but this choice was unfortunate. Survival is certainly critical. Many adaptations of organisms exist because they successfully overcame the forces that impeded survival, such examples are parasites, diseases, food shortages, predators, and Darwin (67) also fashioned the use of term what he believed to be a second evolutionary process, which he called sexual selection. According to sexual selection theory, characteristics that give organisms an advantage in the competition for mates can also evolve. Sexual selection can operate through two processes. The first is intrasexual or same-sex competition. If members of one sex compete with one another, and the victors of these competitions gain preferential sexual access to mates, then these useful qualities lead to success in same-sex competitions will be selected and can evolve over time. These might include athletic ability, social skills to enlist allies, or even a sense of humor that deters a rival. The key point is that whatever qualities lead to success in same-sex competitions can evolve because of the reproductive advantage that accrues to the victor The second component...

Dictatorship essays

Dictatorship essays In modern times, no dictator can take total power by force alone. In order to gain support, they must offer something beneficial to the people. Unfortunately what is thought to be beneficial can be extremely harmful and cruel. Three specific dictators during the World War II period were Mussolini, Hitler, and Stalin. Joseph Stalin was the successor of Lenin after his death. He had only one goal and that was to finish what Lenin had started. Basically that was to build a classless society in which the means of production were in the hands of the people. After Stalin established power, he developed new goals to make the Soviet Union a leader in industry. He wanted to get rid of Russias backwardness. For this reason Stalin proposed several five year plans. These plans were aimed at improving industry and economic growth. He developed a command economy, which meant government made all economic decisions. Under Stalin, the government controlled all businesses. Stalin also took agriculture under control of the government. He forced people to give up their privately owned land and live on government-owned farms or on large farms that were owned and operated by groups, also known as collectives. The state set all prices and controlled access to farm supplies. (Comptons Encyclopedia, 1990, Joseph Sta lin S570) On July 29, 1921, Adolf Hitler was introduced as Fuehrer of the Nazi Party. Fuehrer basically meant leader. By November 1923, the Nazis held approximately 55,000 followers and were the biggest and most powerful in Germany. The Nazi Party demanded action of Hitler. Hitler knew that he would lose his support if he didnt do something fast. So he and his party developed a plan to kidnap the leaders of the Bavarian government and force to accept Hitler as their leader. On November 9th, 1923 Hitler and his Nazis went to Munich and tried to take it over. At this point they were not powerful enough to ...

Near Shoring Assignment Example | Topics and Well Written Essays - 250 words

Near Shoring - Assignment Example This is because near shoring allow businesses to operate near the home country (Shamis, Green, Sorensen t al, 2005). Near shoring allows businesses to conduct their activities in regions with the same time zones. This implies that through this business strategy, it is easier for a business to control all its operations immediately. It also makes the communication process easier. This enables them to coordinate their operations easily and faster. It is difficult for organizations to conducts the same projects in different time zones. In order to achieve the goal of the company, workers in a region with a different time zone have to sacrifice by working at night. Working late in the night may reduce the productivity of workers (McGrath, 2014). Companies that practice near shoring also experience less legal issues compared to those that engage in outsourcing. Outsourcing makes a company to incur huge immigration costs and other costs associated with government regulations. However, in near shoring companies have to only deal with neighboring countries which have less immigration and legal issues. Near shoring also enables a company to enjoy geographical proximity benefits. For instance, it is easier for a company to interact with its customers if they can easily be reached. The situation helps in increasing their

Kinship among South Indian Communities Literature review

Kinship among South Indian Communities - Literature review Example Kinship among South Indian communities has been indispensable and the main form of social organization (caste system). The caste system was a closed hereditary group to which a person belonged strictly by birth. At one point, there were relationships based on endogamous marriage between two people from the same caste. Kinship is also concerned about the productive anxiety of relations of distinction and sameness, the main aspects being the ties that separate or bind. In a fishing village, ‘the Marianad’ what matters is the relations between siblings. The children of the same father and mother, siblings are similar apart from their gender. The strongest differentiation is made between siblings of different gender, a difference that has a great effect in the following generation (Busby 2000; 1995). Therefore, among the â€Å"Marianads† sisters are viewed to be identical in a manner that brother and sister cannot be. Sisters in this tribe, live closely, they are spot ted with each other baby either carrying or feeding it. Contrary, brothers are different in that they move to their wife’s houses in distinct villages, although they view their brother’s children as their own, and they often refer to them as their daughters or sons. The word Dravidian refers to a family dialect mainly spoken in South India. The Dravidian family is different in both origin and structure from the Anglo-Aryan family located in North India. People from South India classify kin based on the difference in sex, the difference in age, the difference in generation, and difference of kin identical with union relationship. This system exemplifies a sociological theory of marriage, and it justifies the issue of someone marrying a cross-cousin (Clark-Deces 2011; Bourdieu 1997). The Marianad people do not have the separate terminologies for the younger and elder sibling, uncles, and aunts. They also do not differentiate between kin identified to ego’s parents via same-sex association (parallel kin) and kin identified to ego’s parents via opposite se-associations. Writers such as Dumont try to suggest the differentiation between the cross and parallel kin in comprehending marriage choices and decisions in South India (Dumont 2006). The children of parents’ same or similar sex siblings (the fathers-in-law and mothers-in-law) are absorbed to the position of elder or young siblings, with whom sexual intimacy, marriage and sexual activities are prohibited. On the other hand, the children of parents’ cross-sex siblings (fathers-in-law and mothers-in-law) are absorbed to the position of spouses or wives with whom marriage is accepted or permitted in that in some castes in south India, it is preferred and prescribed. It is significant to note that these terms recommend separation between relatives (in-laws) and kin, which is not the same as our cultural differentiation between relatives by marriage and blood relatives.