Društvene Mreže - II Kolokvijum

The correct answer suggests that the two terms, "procenat relevantnih informacija koje su ekstrahovane" (percentage of relevant information extracted) and "procenat ekstrahovanih koje su relevantne" (percentage of extracted information that is relevant), are often in conflict with each other. This implies that there is a trade-off between the amount of relevant information that is extracted and the accuracy of the extraction process. In other words, when more information is extracted, there is a higher chance of including irrelevant information, and vice versa.

This answer suggests that Person A may have a high level of trust in Person B as a doctor, but does not trust them as a driver. This implies that Person A's trust is specific to the topic or area of expertise, rather than trusting Person B as a whole.

Amazon je poznat po tome što je uspešno iskoristio long tail u sistemu preporuka. Long tail se odnosi na strategiju koja se fokusira na prodaju velikog broja manje popularnih proizvoda, umesto samo na najpopularnije. Amazon je uspeo da iskoristi ovu strategiju tako što je omogućio kupcima da pronađu i kupuju proizvode koji su specifični za njihove interese, čak i ako ti proizvodi nisu široko poznati ili popularni. Ovo je doprinelo povećanju prodaje i zadovoljstva korisnika, što je jedan od razloga zašto je Amazon postao jedan od najvećih online trgovaca.

The filter bubble problem refers to the isolation of users from content that is outside their expressed sphere of interest. This occurs when algorithms and personalized recommendations only show users content that aligns with their previous preferences and interests, creating a narrow and limited view of the world. This can lead to a lack of exposure to diverse perspectives and information, limiting users' ability to access a wide range of content and potentially reinforcing existing biases and beliefs.

Poverenje jedne strane u drugu ne znači obavezno i obrnuto, što znači da je poverenje asimetrično. Ovo znači da jedna strana može imati poverenje u drugu stranu, ali ta druga strana ne mora nužno imati isto poverenje u prvu stranu. Asimetrično poverenje se često javlja u odnosima gde postoji razlika u moći, autoritetu ili ranijim iskustvima između strana.

The correct answer is "Opisuju međusobne veze između tri čvora." This answer is supported by the given information that "Simelijeve veze" describe the mutual connections between three nodes.

When two separate clusters have non-redundant information, it leads to the formation of structural holes.

Poverenje kao rezultat saradnje zasniva se na prethodnom iskustvu. To znači da se poverenje gradi na osnovu prethodnih interakcija i iskustava sa osobom ili organizacijom. Kada smo imali pozitivna iskustva u saradnji sa nekim, skloni smo da im više verujemo u budućim situacijama. Ovo je važan preduslov za uspostavljanje poverenja u bilo kojoj vrsti odnosa, jer nam prethodno iskustvo pruža informacije o tome koliko možemo verovati drugoj strani.

Crowdsourcing allows companies to explore and investigate problems without incurring any costs. By leveraging the collective intelligence and diverse perspectives of a crowd, companies can gather valuable insights and solutions without the need for expensive research or development. This can be particularly advantageous as it allows companies to tap into a wide range of knowledge and expertise without the limitations of time or budget constraints. Additionally, crowdsourcing can help in generating innovative ideas and solutions that may not have been possible through traditional methods.

The correct answer is "Najduža najkraća putanja između bilo koja 2 čvora u mreži". This means that it is the longest shortest path between any two nodes in the network, indicating the maximum distance that can be traveled while still taking the shortest route. It implies that there may be multiple paths between two nodes, but the longest shortest path is the one being referred to here.

The correct answer is 2 because the question is asking about the number of neural networks in YouTube's recommendation system. The answer "2" indicates that the system consists of two neural networks.

Vikipedija je primer mudrosti mase jer je to online enciklopedija koja se stvara i uređuje kolektivnim naporom velikog broja ljudi. Svako može doprineti i uređivati članke na Vikipediji, što omogućava da se koristi znanje i iskustvo šireg kruga ljudi. Ova praksa se naziva crowdsourcing, gde se koristi kolektivna inteligencija i znanje mase kako bi se stvorila kvalitetna i sveobuhvatna enciklopedija. Takođe, folksonomija se odnosi na klasifikaciju i organizaciju informacija na osnovu oznaka koje korisnici sami dodeljuju, što takođe može biti primenjeno na Vikipediju.

The correct answer is "Da preporuči sadržaj iz long tail-a." This means that the quality of a recommendation system is measured by its ability to recommend content from the long tail. The long tail refers to a large number of unique items that have low popularity or demand individually but collectively make up a significant portion of the market. A good recommendation system should be able to suggest niche or less popular content to users, not just focus on popular or mainstream items.

The correct answer is "Sve što važi za poverenje u opštem slučaju važi i za poverenje na vebu". This means that everything that applies to trust in general also applies to trust on the web. This suggests that the principles and factors that contribute to trust in offline interactions also apply to online interactions. Trust is an important aspect of any relationship or interaction, whether it is offline or on the web.

Pod društvenim kapitalom se podrazumeva kapital čija je osnova u društvenim odnosima, a koji doprinosi privrednom i socijalnom razvoju društva. Ovo znači da društveni kapital obuhvata resurse kao što su mreže socijalnih veza, poverenje, recipročnost i saradnja među ljudima i organizacijama. Kroz ove društvene veze i resurse, društveni kapital može pozitivno uticati na ekonomski rast i razvoj zajednice, kao i na socijalnu koheziju i blagostanje.

The correct answer suggests that network closure occurs when there are strong connections between actors. This means that the individuals or entities within the network have strong ties or relationships with each other, which contributes to the overall cohesion and interconnectedness of the network. Strong connections between actors can lead to increased collaboration, information sharing, and mutual support within the network, ultimately resulting in network closure.

Crowdsourcing involves outsourcing tasks to a large group of people, typically through an online platform. One drawback of crowdsourcing is the lack of control over the development or final product. Since the tasks are delegated to a diverse group of individuals, there is a possibility that the end result may not meet the desired standards or expectations. This lack of control can lead to inconsistencies in the quality, design, or functionality of the product or solution being crowdsourced.

Not all types of media are suitable for every type of content in implementing social learning.

The correct answer suggests that real (online) social networks are characterized by the presence of a large number of small, well-connected groups. This means that while there may be many small groups within the network, the members within each group are closely connected to each other. This is in contrast to options such as a small number of large, loosely connected groups or a large number of small, loosely connected groups.

Transitivity, along with homophily, leads to the formation of cliques.

The closeness centrality measures the proximity of a node to other nodes in a network. It calculates the average shortest path length from the node to all other nodes in the network. Therefore, the correct answer suggests that closeness centrality represents the proximity of the observed node to the central position in the network.

Preferential attachment is a process in network growth where the majority of new nodes connect to existing nodes that already have a high degree. This means that popular nodes with many connections are more likely to attract new connections, leading to the rich-get-richer phenomenon in network formation. This process can explain the growth of certain social networks or the distribution of links on the internet, where highly connected nodes tend to become even more connected over time.

Crowdsourcing is the practice of obtaining necessary services, ideas, or content by seeking contributions from a wide range of people, especially online communities, rather than through traditional means such as employees.

The correct answer is "Baze znanja" because Facebook's Entity Graph can be seen as a knowledge base. It is a database that contains information about entities (such as people, places, and things) and their relationships. This knowledge base helps Facebook understand and organize the vast amount of data it collects from its users, allowing for more targeted advertising, personalized recommendations, and improved search functionality.

Apsolutno poverenje refers to complete trust where the outcome is fully known before the action is taken. This means that there is no uncertainty or risk involved, as the result is already predetermined or guaranteed. It is not recommended because in most situations, it is unrealistic to have absolute certainty about the outcome before any action is taken. Trust is usually based on previous experience or a necessary condition of collaboration, where there is some level of uncertainty or unknown factors involved.

The given answer, "Merljivo" (measurable), suggests that trust can be quantified or measured using various criteria. This implies that trust is not an abstract concept but can be evaluated and assessed based on specific factors or indicators. This aligns with the idea that trust can be analyzed and understood through observable and measurable behaviors, actions, or outcomes.

Long tail se odnosi na veliki deo informacija na internetu za koje mali broj korisnika pokazuje interesovanje. Ovo se odnosi na manje popularne, specifične ili nišne sadržaje koji imaju manji broj poseta ili pregleda u odnosu na popularne sadržaje koji privlače veći broj korisnika. Ova terminologija se često koristi u kontekstu pretraživačkih sistema i e-trgovine, gde se ističe važnost dugog repa odnosno manje popularnih proizvoda ili usluga koji zajedno mogu činiti značajan deo ukupnog prometa.

The FOAF (Friend of a Friend) model of trust is characterized by identifying users through their unique email addresses. This means that each user is associated with a specific email address, allowing for the establishment of trust relationships within the FOAF model. This model relies on the concept of social networks and the connections between individuals, using email addresses as a means of identification and verification. Therefore, the correct answer is FOAF model poverenja.

not-available-via-ai

The correct answer is "Poverenje u celini povlači poverenje i u temu i u osobu". This means that trust in a person as a whole includes trust in both the topic and the person themselves. In other words, if someone has trust in someone, it also implies trust in the topic they are discussing.

The correct answer is "Avokado metrike." This is because the sentence mentions that good nodes on a social network rarely certify bad nodes, while bad nodes certify each other. The term "Avokado metrike" is not familiar, but it can be inferred that it refers to a metric or measurement related to the certification or evaluation of nodes on a social network.

If a trust edge is added to the trust network graph from one node to another without changing the trust between those two nodes, it indicates that the trust has strong overall invariance. This means that the trust remains unchanged regardless of the additional trust edge, suggesting a high level of trust stability and consistency in the network.

The correct answer is "Broj čvorova u mreži" which translates to "Number of nodes in the network" in English. This means that the size of the network refers to the number of nodes or vertices present in the network. The size of a network does not refer to the number of links or connections, the diameter of the network, or the ratio of nodes to links.

The given answer suggests that diversity of thinking, independence, decentralization, and aggregation are necessary conditions for a crowd to be wise. This implies that in order for a group to make wise decisions or come up with intelligent solutions, it is important to have a variety of perspectives, individual autonomy, a decentralized decision-making process, and the ability to combine and analyze different opinions.

The correct answer states that nodes in a graph represent entities between which we observe connections. This means that the nodes in a graph represent the entities or objects of interest, and the connections between the nodes represent the relationships or interactions between these entities. This is a common way to interpret and analyze graphs in various fields, including social networks.

The correct answer is "Sistem preporuke treba da preporuči nešto što je korisniku nepoznato a sistem pretrage da locira ono što je poznato." This is because the purpose of a recommendation system is to suggest something unknown to the user based on their preferences or previous interactions. On the other hand, a search system is designed to find and locate something that is already known or requested by the user. Therefore, the recommendation system should provide new and unfamiliar content, while the search system should help locate familiar content.

The correct answer is "Koji pokazuju identično ponašanje na istu dinamiku ponašanja." This answer suggests that similar users are those who exhibit identical behavior patterns on the same behavioral dynamics. This implies that they have similar usage patterns and engage with the web services in a similar manner.

Društveni kapital se odnosi na resurse koje pojedinci ili grupe dobijaju kroz održavanje mreže društvenih veza. Ovaj termin se koristi za opisivanje koristi koje proizlaze iz veza i povezanosti sa drugim ljudima u društvu, kao što su podrška, informacije, pristup resursima i mogućnosti. Društveni kapital može biti vredan resurs za pojedince i grupe jer im omogućava da ostvare svoje ciljeve i poboljšaju svoj društveni status.

X.509 and PGP (Pretty Good Privacy) are both cryptographic protocols used for establishing and verifying the authenticity of digital certificates. These certificates are used to validate the identity of individuals or entities in online communication. Therefore, X.509 and PGP can be seen as models of trust or trust frameworks that provide a basis for establishing trust and ensuring the security of digital communication. They define the rules and procedures for issuing, managing, and validating certificates, thus forming the foundation of trust in online transactions and communications.

Kolektivna inteligencija se odnosi na inteligenciju koja se razvija kroz saradnju i takmičenje više pojedinaca. Ova vrsta inteligencije omogućava grupi da rešava probleme, donosi odluke i stvara inovativna rešenja na efikasan način. Kroz zajednički rad i razmenu ideja, kolektivna inteligencija može biti veća od inteligencije pojedinaca koji je čine, što je osnova za uspeh koncepta "mudrosti gomile" ili "wisdom of crowds". Crowdsourcing i folksonomija su srodni koncepti, ali se odnose na druge aspekte kolektivne inteligencije.

The Pirsonov koeficijent korelacije, or Pearson correlation coefficient, is used to measure the similarity between users or products. It quantifies the strength and direction of the linear relationship between two variables. It is commonly used in recommendation systems to determine the similarity between users or products based on their preferences or characteristics. By calculating the Pearson correlation coefficient, the system can identify similar users or products and make recommendations based on those similarities.

The correct answer is "Analiza sentimenata" which translates to "Sentiment analysis" in English. This refers to the process of analyzing messages exchanged over social media in order to identify the attitude, opinion, or emotion expressed in them. Sentiment analysis is a valuable tool for businesses and organizations to understand public opinion, customer feedback, and brand reputation. It involves using natural language processing and machine learning techniques to classify texts as positive, negative, or neutral, providing valuable insights for decision-making and strategy development.

The correct answer is "Kolektivne inteligencije" (Collective intelligence). Google Page Rank is an example of collective intelligence because it uses the collective knowledge and opinions of internet users to determine the relevance and importance of web pages. The algorithm analyzes the links between pages and assigns a rank based on the collective intelligence of the online community.

Poverenje na društvenim mrežama se sastoji od tri komponente.

Baze znanja dostupne na Web-u (Knowledge bases available on the Web) are used for uniquely identifying entities. These knowledge bases contain information that can be accessed by anyone on the internet, making them a valuable resource for identifying and obtaining information about entities. Unlike the other options listed, which are limited to specific organizations or require access to protected servers, knowledge bases available on the Web provide a wide range of information that can be used for identification purposes.

The correct answer is "Preporučiti korisniku ono što odgovara njegovim potrebama." This answer states that recommendations should be based on the user's needs. This means that the recommendations should be tailored to the individual user and should provide them with content or products that are relevant and useful to them.

Baze znanja dostupne na Web-u (Knowledge bases available on the Web) are used for unique identification of entities. These knowledge bases provide a wide range of information and resources that can be accessed by anyone with an internet connection. They serve as a valuable source of information for identifying and understanding various entities, such as individuals, organizations, concepts, etc. These knowledge bases are accessible to a global audience and can be utilized for various purposes, including research, education, and decision-making.

When multiple users tag a specific content with different terms from different dictionaries, creating multiple metadata for the content, it represents a broader folksonomy.

The correct answer is "Broj stvarnih veza u mreži / ukupan broj mogućih veza u mreži". This is because the density of a network is defined as the ratio of the actual number of connections in the network to the total number of possible connections in the network.

By providing explanations along with recommendations, trust in the recommendation system is encouraged. This helps users understand why certain recommendations are being made and increases transparency in the system. Users are more likely to trust and follow recommendations when they have a clear understanding of the reasoning behind them.

Interpretacija betweenness centrality u kontekstu društvenih mreža se odnosi na verovatnoću da je data osoba posrednik između neke druge dve osobe iz date mreže. Ova mera meri koliko često se određeni čvor nalazi na najkraćim putevima između ostalih čvorova u mreži. Ako je neka osoba često na putu između drugih ljudi, to ukazuje na to da ima veću verovatnoću da je posrednik u komunikaciji ili prenosu informacija između tih ljudi.

The correct answer is "Metrika poverenja." This is because the question is asking for the discipline that requires an initial level of trust in an action for which there is no previous knowledge, experience, and outcomes. "Metrika poverenja" refers to the measurement or metric of trust, indicating that this discipline involves quantifying or measuring trust in such situations. The other options, "Skala poverenja" (trust scale), "Mreža poverenja" (trust network), and "Model poverenja" (trust model), do not specifically address the aspect of measuring trust in the given context.

The correct answer is "Izolacije korisnika od sadržaja koji ne pripadaju sferi njegovog iskazanog interesovanja." This answer explains that the filter bubble problem occurs when users are isolated from content that does not align with their expressed interests. This means that users are only exposed to information and articles that match their existing preferences, which can limit their exposure to diverse perspectives and potentially reinforce their existing beliefs.

Explicit rating allows users to clearly indicate their taste/opinion/view, which helps in understanding their preferences. This information can be used to personalize recommendations and improve user experience. Additionally, explicit rating encourages trust in the system as users feel their opinions are valued and taken into account. It also enables continuous and effortless data collection without burdening the users. Moreover, users tend to enjoy rating articles/content, which can lead to increased engagement and participation.

The given correct answer states that the assumption that user interests do not change over time is a heuristic used in recommendation systems. This means that recommendation systems make the assumption that user preferences and interests remain constant, which helps in simplifying the recommendation process. However, it is important to note that this assumption is not accurate in reality, as user interests are likely to change over time. Nonetheless, this assumption has been found to be irrelevant in the context of recommendation systems.

Social learning refers to learning that takes place in a social context on the internet. It involves the use of online platforms and websites where individuals can interact, share knowledge, and collaborate with others. This type of learning is not limited to social media sites but can occur in various online environments where people engage in discussions, participate in online communities, and learn from each other's experiences and expertise. Therefore, the correct answer is "U društvenom kontekstu na Internetu" (In a social context on the internet).

The correct answer is "Postoji nedostatak parova (korisnik, ocena)". This means that there is a lack of data where both the user and their rating for a particular product or content are missing. This sparsity problem occurs when there are incomplete or missing ratings, making it difficult to analyze or make predictions based on user preferences or similarities between users or items.

Betweenness centrality measures the number of nodes between which the observed node lies. It quantifies the extent to which a node serves as a bridge or intermediary between other nodes in a network. Nodes with high betweenness centrality can control the flow of information or resources between different parts of the network, making them influential in the network structure.

Folksonomija je naziv za servis koji omogućava obeležavanje sadržaja na Web-u od strane svakog pojedinca. Ova praksa omogućava korisnicima da dodaju oznake (tagove) sadržaju kako bi ga organizovali, pretraživali i delili sa drugima. Ove oznake su obično reči ili kratke fraze koje opisuju sadržaj i omogućavaju korisnicima da ga lakše pronađu. Folksonomija je demokratski pristup kategorizaciji i organizaciji sadržaja na Web-u, jer je svaki pojedinac slobodan da dodaje oznake prema sopstvenom razumevanju i interesovanjima.

The correct answer is "Za semantičko indeksiranje koriste se baze znanja koje se nalaze na Web-u." This statement explains that semantic indexing utilizes knowledge bases found on the web. Semantic indexing involves extracting information and determining the correct meaning of entities in a given context, but it is not the same as information retrieval or searching for information.

The correct answer is "Nedostaju informacije o novim korisnicima/artiklima" which means "There is a lack of information about new users/items". This is a possible explanation for the cold start problem, as it occurs when there is insufficient data available about new users or items. Without enough information, it becomes difficult to make recommendations or determine similarities between users or items. Additionally, if there are no direct ratings or reviews for a product/content, it further adds to the cold start problem.

The given options "IBM Watson and Wolfram Alpha are examples of:" do not clearly indicate what they are examples of. Therefore, an explanation cannot be provided as the question is incomplete and not readable.

The correct answer is "Kompozitno" because the statement mentions that trust is formed based on a larger number of recommendations from others. This suggests that trust is a composite phenomenon, meaning it is formed by combining multiple factors or elements.

The explanation for the correct answer is that if there is trust in another party, it usually implies trust in the recommendations made by that party. This suggests that trust is transitive, meaning that it can be passed on from one entity to another.

Ekstrakcija informacija i pretraživanje informacija su slični procesi, ali imaju različite funkcije. Ekstrakcija informacija analizira tekstove i pronalazi relevantne segmente teksta koje zatim predstavlja. Sistem za pretraživanje, s druge strane, pronalazi relevantne tekstove na osnovu upita ili ključnih reči. Dakle, ekstrakcija informacija se fokusira na pronalaženje relevantnih segmenata unutar teksta, dok sistem za pretraživanje se fokusira na pronalaženje relevantnih celih tekstova.

If removing a node from the trust network graph and redirecting the edges to its trusted nodes does not change the overall trust, then it indicates that the trust has weak general invariance. This means that the trust relationships within the network are not strongly influenced by the presence or absence of a specific node.

The explanation for the given correct answer is that the statement "Ovaj blog post je napisao/preporučio tvoj prijatelj X" indicates that the recommendation is based on social connections. This suggests that the system is using a method that relies on the connections and interactions between friends or acquaintances to make recommendations.

The correct answer is "Se smatra prednošću za odlučivanje mudrih gomila". This means that forming a trust network without a central decision-making system is considered an advantage for wise crowds. This suggests that decentralized decision-making allows for more independent and diverse perspectives, leading to better collective decisions. However, it is important to note that this advantage may only apply to small-scale crowds.

The system for searching recognizes entities in the text and uniquely identifies them. It analyzes texts and presents only segments of information (extracted from the text) that the user may be interested in. It finds (potentially) relevant texts and presents them to the user. It determines what the entity refers to.

The correct answer is FOAF modelom. FOAF (Friend of a Friend) is a model that allows the connection of statements about which user knows whom and how much trust they have in the users they know. It is a way to represent social networks and relationships between individuals on the web. It provides a framework for expressing relationships, such as friendship, trust, and acquaintance, using RDF (Resource Description Framework) data model.

The correct answer is "Broj stvarnih veza u mreži / ukupan broj čvorova u mreži". This is because the metric "links per node" is defined as the ratio of the number of actual connections in the network to the total number of nodes in the network.

The correct answer suggests that the advantage of weakly connected networks is that the transfer of information and new ideas is much more intensive. This means that in a network where person A is weakly connected to persons B and C, it is highly likely that persons B and C are also connected to each other. This intensifies the exchange of information and new ideas within the network.

The interpretation of eigenvector centrality in the context of social networks is the extent to which a person is connected to influential or well-connected members of the network. This measure considers not only the number of direct connections a person has but also the importance of those connections. It takes into account the influence and connectivity of the individuals that a person is connected to, rather than just the number of connections they have.

The given correct answer suggests that the method being applied in the given scenario is the "Nearest Neighbors" method. This method involves recommending a product Y to users who have purchased product X based on the preferences and buying patterns of their closest neighbors or similar users. It does not rely on social connections or knowledge base, but rather on the similarity of users' preferences and past behaviors.

The correct answer is "Je generisan slabim vezama" which means "It is generated by weak ties". This explanation suggests that social capital is formed through connections with people who are in different social positions. Weak ties refer to relationships with acquaintances or distant connections, rather than close friends or family members. These weak ties can provide access to new information, resources, and opportunities, contributing to the generation of social capital.

The correct answer is "Upotrebu tehnologije za generisanje društvenog kapitala." This means that social-technical capital refers to the use of technology to generate social capital. It implies that technology can be utilized to enhance social connections, networks, and relationships, leading to the creation of social capital.

The given correct answer suggests that the explanation "Ovaj novinski članak govori o temi X koja pripada skupu tema za koje ste ranije pokazali interesovanje" is obtained when the system applies the method that is based on content. This means that the system is using the content of the article to determine the topic and relevance to the user's previous interests. It does not rely on knowledge bases, social connections, or nearest neighbors for this particular explanation.

In the context of information extraction, the most challenging task is the recognition of events. This involves identifying and extracting information about specific occurrences or happenings from a given text or dataset. It requires understanding the context, identifying relevant keywords or phrases, and determining the relationships between different elements in order to accurately recognize and extract event-related information. This task can be particularly difficult due to the complexity and variability of events, as well as the potential ambiguity in language and expression.

Degree centrality measures the number of nodes directly connected to a given node. It quantifies the importance or influence of a node based on its connections to other nodes in the network. In other words, it represents how many nodes are directly linked to the observed node.

This metric is used when the lengths of the shortest paths in the network significantly differ. It helps to quantify and compare the average length of the shortest paths between any two nodes in the network. By calculating this metric, we can determine the extent to which the lengths of the shortest paths vary in the network, which can provide insights into the network's structure and connectivity.

Semantic indexing is a technique used to uniquely name entities. In this context, the use of knowledge bases makes sense as they contain structured information about various entities. These knowledge bases can be used to create semantic indexes that provide unique names or identifiers for entities. Therefore, the correct answer is "Baze znanja" (Knowledge bases).

Recommendations based on content use data from the user's profile, contextual parameters, and product data. This means that the recommendations take into account information about the user, such as their preferences and past behavior, as well as the context in which the recommendations are being made, and the specific details of the products being recommended. This comprehensive approach ensures that the recommendations are personalized and relevant to the user's interests and needs.

The interpretation of degree centrality in the context of social networks is the number of other people that a given person can directly contact. Degree centrality measures the number of direct connections a node has in a network, indicating how well connected and accessible that person is within the network.

Poslovne analitike se odnose na ekstrakciju entiteta iz novinskih članaka, blog postova i slično. To znači da se koristi analiza teksta kako bi se identifikovali relevantni entiteti, kao što su imena, lokacije, događaji ili bilo koje druge informacije koje su bitne za poslovne svrhe. Ova vrsta analize može pomoći u donošenju informisanih odluka i razumevanju trendova i potreba u poslovnom okruženju.

Poverenje je relevantno jer utiče na dalje akcije korisnika. Kada korisnik ima poverenje u nešto, verovatnije je da će preduzeti određene akcije u vezi sa tim. Poverenje je važno za uspostavljanje odnosa i održavanje saradnje između ljudi ili organizacija. Bez poverenja, teško je ostvariti uspeh u bilo kojoj vrsti interakcije. Zbog toga je poverenje relevantno jer ima direktnu vezu sa onim što ljudi odlučuju da urade.

Interpretacija blizine čvora (closeness centrality) u kontekstu društvenih mreža se odnosi na koliko brzo data osoba može da dopre do bilo koje druge osobe u mreži. Ova mera meri prosečnu udaljenost između datog čvora i svih ostalih čvorova u mreži. Što je manja udaljenost, to je veća blizina čvora. Ova interpretacija se fokusira na efikasnost komunikacije i dostupnost informacija u mreži.

The correct answer is "Metodu kolaborativnog filtriranja koju koriste sistemi preporuka." This answer states that the method of collaborative filtering is used by recommendation systems. Collaborative filtering is a technique that involves making recommendations based on the preferences and behavior of similar users. It analyzes the past actions and patterns of users to make predictions and suggest items that they may like. This method is widely used in recommendation systems to provide personalized recommendations to users based on the input and feedback of other users.

Povezujući društveni kapital je kapital koji nastaje povezivanjem ljudi koji se nalaze na različitim društvenim položajima. Ova vrsta kapitala se generiše kroz jake veze, što znači da su veze između ljudi dublje i intenzivnije. To omogućava razmenu resursa, informacija i podrške među povezanim pojedincima, što može biti od velike koristi za sve uključene.

The correct answer is "Je kapital koji nastaje povezivanjem ljudi koji se nalaze na različitim društvenim položajima." This means that social capital is generated by connecting people who are in different social positions. This suggests that having connections with individuals from diverse backgrounds and social positions can contribute to the development of social capital.

The correct answer is "Mera preciznosti sistema za ekstrakciju informacija" (Measure of precision of the information extraction system). Precision is a measure of the accuracy of the system in extracting relevant information. It indicates the proportion of correctly extracted information out of all the information extracted. It is an important metric in evaluating the performance of information extraction systems as it focuses on the correctness of the extracted information rather than the total amount of information extracted.

Neformalne asocijacije se smatraju bitnim mehanizmom stvaranja društvenog kapitala.

The correct answer is "Određivanje na šta se odnosi entitet" because recognizing named entities in information extraction involves determining what the entity refers to. This means identifying the specific entity or object that is being mentioned or referred to in the given text.

The correct answer is "Određivanje tipičnog prideva entiteta" which means determining the typical adjective of the entity. This suggests that the task is to identify or describe the characteristic adjective that best describes the entity in question.

The correct answer explains that the system for extraction analyzes texts and presents only segments of information that are extracted from the text and may be of interest to the user. This means that the system filters out irrelevant information and focuses on providing the user with the most relevant and important information from the texts.

The correct answer is "Određivanje na šta se odnosi entitet" because resolving coreferences in information extraction involves determining what entity a pronoun or noun phrase refers to. This process helps in understanding the context and correctly extracting relevant information.

The given correct answer states that "Prepoznavanje relacija koje postoje između (imenovanim) entitetima je zadatak ekstrakcije informacija." This means that recognizing the relationships between named entities is a task of information extraction. It implies that the process involves identifying and extracting relevant information from the entities in order to understand the connections or associations between them. This task is not related to analyzing segments or content, but it is an important factor in determining the performance of content analysis systems.

Personalized recommendations use only the data from the user's profile and contextual parameters. This means that the recommendations are based on the user's personal information and the current context in which they are using the system. It does not take into account data about other users or data from the internet about the user or products.

Identifikacija ključnih tema/koncepata i semantičko indeksiranje su slični termini, jer kod oba procesa se teme/koncepti ne vezuju za pojedinačne reči. Identifikacija ključnih tema/koncepata se odnosi na pronalaženje glavnih tema ili ideja u tekstu, dok se semantičko indeksiranje odnosi na povezivanje tema/koncepata sa sličnim terminima ili pojmovima. Oba procesa se fokusiraju na suštinske elemente teksta, a ne na pojedinačne reči.

After identifying the key themes/concepts, the texts are analyzed and only segments of information are presented. In this process, the texts are usually ranked according to their importance or significance. This helps in prioritizing the information and presenting the most relevant and significant texts to the user.