This diploma thesis deals with problems of development of chatbots. The theoretical part of the thesis introduces the concept of the conversational interface in general and analyzes available technologies for its development. The practical part deals with the design and implementation of a particular chatbot, whose goal is to be a virtual assistant in the process of selecting and purchasing goods. This is accomplished by connecting the chatbot to the product information management system using OData web services. One of the biggest problems was to determine the order of questions asked about product properties. For the implementation was used decision tree theory.

Product information search has become one of the most important application areas of the Web. Especially considering pricey technical products, consumers tend to carry out intensive research activities previous to the actual acquisition for creating an all-embracing view on the product of interest. Federated search backed by ontology-based product information representation shows great promise for easing this research process. The topic of this thesis is to develop a comprehensive technique for locating, extracting, and integrating information of arbitrary technical products in a widely unsupervised manner. The resulting homogeneous information sets allow a potential consumer to effectively compare technical products based on an appropriate federated product information system.:1. Introduction 1.1. Online Product Information Research 1.1.1. Current Online Product Information Research 1.1.2. Aspired Online Product Information Research 1.2. Federated Shopping Portals 1.3. Research Questions 1.4. Approach and Theses 1.4.1. Approach 1.4.2. Theses 1.4.3. Requirements 1.5. Goals and Non-Goals 1.5.1. Goals 1.5.2. Non-Goals 1.6. Contributions 1.7. Structure 2. Federated Information Systems 2.1. Information Access 2.1.1. Document Retrieval 2.1.2. Federated Search 2.1.3. Federated Ranking 2.2. Information Extraction 2.2.1. Information Extraction from Structured Sources 2.2.2. Information Extraction from Unstructured Sources 2.2.3. Information Extraction from Semi-structured Sources 2.3. Information Integration 2.3.1. Ontologies 2.3.2. Ontology Matching 2.4. Information Presentation 2.5. Product Information 2.5.1. Product Information Source Characteristics 2.5.2. Product Information Source Types 2.5.3. Product Information Integration Types 2.5.4. Product Information Types 2.6. Conclusions 3. A Federated Product Information System 3.1. Finding Basic Product Information 3.2. Enriching Product Information 3.3. Administrating Product Information 3.4. Displaying Product Information 3.5. Conclusions 4. Product Information Extraction from the Web 4.1. Vendor Product Information Search 4.1.1. Vendor Product Information Ranking 4.1.2. Vendor Product Information Extraction 4.2. Producer Product Information Search 4.2.1. Producer Product Document Retrieval 4.2.2. Producer Product Information Extraction 4.3. Third-Party Product Information Search 4.4. Conclusions 5. Product Information Integration for the Web 5.1. Product Representation 5.1.1. Domain Product Ontology 5.1.2. Application Product Ontology 5.1.3. Product Ontology Management 5.2. Product Categorization 5.3. Product Specifications Matching 5.3.1. General Procedure 5.3.2. Elementary Matchers 5.3.3. Evolutionary Matcher 5.3.4. Naïve Bayes Matcher 5.3.5. Result Selection 5.4. Product Specifications Normalization 5.4.1. Product Specifications Atomization 5.4.2. Product Specifications Value Normalization 5.5. Product Comparison 5.6. Conclusions 6. Evaluation 6.1. Implementation 6.1.1. Offers Service 6.1.2. Products Service 6.1.3. Snippets Service 6.1.4. Fedseeko 6.1.5. Fedseeko Browser Plugin 6.1.6. Fedseeko Mobile 6.1.7. Lessons Learned 6.2. Evaluation 6.2.1. Evaluation Measures 6.2.2. Gold Standard 6.2.3. Product Document Retrieval 6.2.4. Product Specifications Extraction 6.2.5. Product Specifications Matching 6.2.6. Comparison with Competitors 6.3. Conclusions 7. Conclusions and Future Work 7.1. Summary 7.2. Conclusions 7.3. Future Work A. Pseudo Code and Extraction Properties A.1. Pseudo Code A.2. Extraction Algorithm Properties A.2.1. Clustering Properties A.2.2. Purging Properties A.2.3. Dropping Properties B. Fedseeko Screenshots B.1. Offer Search B.2. Product Comparison / Die Produktinformationssuche hat sich zu einem der bedeutendsten Themen im Web entwickelt. Speziell im Bereich kostenintensiver technischer Produkte führen potenzielle Konsumenten vor dem eigentlichen Kauf des Produkts langwierige Recherchen durch um einen umfassenden Überblick für das Produkt von Interesse zu erlangen. Die föderierte Suche in Kombination mit ontologiebasierter Produktinformationsrepräsentation stellt eine mögliche Lösung dieser Problemstellung dar. Diese Dissertation stellt Techniken vor, die das automatische Lokalisieren, Extrahieren und Integrieren von Informationen für beliebige technische Produkte ermöglichen. Die resultierenden homogenen Produktinformationen erlauben einem potenziellen Konsumenten, zugehörige Produkte effektiv über ein föderiertes Produktinformationssystem zu vergleichen.:1. Introduction 1.1. Online Product Information Research 1.1.1. Current Online Product Information Research 1.1.2. Aspired Online Product Information Research 1.2. Federated Shopping Portals 1.3. Research Questions 1.4. Approach and Theses 1.4.1. Approach 1.4.2. Theses 1.4.3. Requirements 1.5. Goals and Non-Goals 1.5.1. Goals 1.5.2. Non-Goals 1.6. Contributions 1.7. Structure 2. Federated Information Systems 2.1. Information Access 2.1.1. Document Retrieval 2.1.2. Federated Search 2.1.3. Federated Ranking 2.2. Information Extraction 2.2.1. Information Extraction from Structured Sources 2.2.2. Information Extraction from Unstructured Sources 2.2.3. Information Extraction from Semi-structured Sources 2.3. Information Integration 2.3.1. Ontologies 2.3.2. Ontology Matching 2.4. Information Presentation 2.5. Product Information 2.5.1. Product Information Source Characteristics 2.5.2. Product Information Source Types 2.5.3. Product Information Integration Types 2.5.4. Product Information Types 2.6. Conclusions 3. A Federated Product Information System 3.1. Finding Basic Product Information 3.2. Enriching Product Information 3.3. Administrating Product Information 3.4. Displaying Product Information 3.5. Conclusions 4. Product Information Extraction from the Web 4.1. Vendor Product Information Search 4.1.1. Vendor Product Information Ranking 4.1.2. Vendor Product Information Extraction 4.2. Producer Product Information Search 4.2.1. Producer Product Document Retrieval 4.2.2. Producer Product Information Extraction 4.3. Third-Party Product Information Search 4.4. Conclusions 5. Product Information Integration for the Web 5.1. Product Representation 5.1.1. Domain Product Ontology 5.1.2. Application Product Ontology 5.1.3. Product Ontology Management 5.2. Product Categorization 5.3. Product Specifications Matching 5.3.1. General Procedure 5.3.2. Elementary Matchers 5.3.3. Evolutionary Matcher 5.3.4. Naïve Bayes Matcher 5.3.5. Result Selection 5.4. Product Specifications Normalization 5.4.1. Product Specifications Atomization 5.4.2. Product Specifications Value Normalization 5.5. Product Comparison 5.6. Conclusions 6. Evaluation 6.1. Implementation 6.1.1. Offers Service 6.1.2. Products Service 6.1.3. Snippets Service 6.1.4. Fedseeko 6.1.5. Fedseeko Browser Plugin 6.1.6. Fedseeko Mobile 6.1.7. Lessons Learned 6.2. Evaluation 6.2.1. Evaluation Measures 6.2.2. Gold Standard 6.2.3. Product Document Retrieval 6.2.4. Product Specifications Extraction 6.2.5. Product Specifications Matching 6.2.6. Comparison with Competitors 6.3. Conclusions 7. Conclusions and Future Work 7.1. Summary 7.2. Conclusions 7.3. Future Work A. Pseudo Code and Extraction Properties A.1. Pseudo Code A.2. Extraction Algorithm Properties A.2.1. Clustering Properties A.2.2. Purging Properties A.2.3. Dropping Properties B. Fedseeko Screenshots B.1. Offer Search B.2. Product Comparison

Syfte: Studien syftar till att förstå hur hållbara produktattribut kan verka i samklang eller kombineras med produktattribut som tillfredsställer konsumentens egennytta. Det möjliggörs genom kartläggning av vilken produktinformation om hållbara produktattribut som svenska modekonsumenter efterfrågar i digitala sammanhang för att kunna göra miljömedvetna val, och som då kan bidra med positiva köpintentioner. Detta för att kunna minska avståndet mellan intresse och handling vad gäller hållbar konsumtion. Metod: Metodansatsen som användes var induktiv och för datainsamling användes blandad metod. Litteratursökning följdes av en kvantitativ webbaserad enkät, vilken kompletterades med kvalitativa, semistrukturerade intervjuer. Urvalet för både enkät och intervjuer var svenska modekonsumenter i åldrarna 18–35. Kvantitativa och kvalitativa data jämfördes och analyserades sedan mot teoretiska perspektiv om att detaljerad produktinformation kan fungera som substitut till fysisk interaktion och att konsumenter inte prioriterar hållbara produktattribut som inte uppfyller egennytta. Resultaten diskuterades och kunde slutligen presenteras som slutsatser. Studien fokuserade endast på modekonsumtion i en online kontext. Den avsåg heller inte att se till vilka hållbara produktattribut som är bättre eller sämre ur ett hållbarhetsperspektiv. Resultat: Resultaten visar att produktattribut som kan anses tillfredsställa egennytta prioriteras framför hållbara produktattribut vid konsumtion av mode, men att hållbar information kring modeprodukter ökar köpintentioner. Konsumenter upplever dock att företag inte kommunicerar tillräckligt kring hållbarhet och efterfrågar transparens och mer detaljerad information. Resultatet visar en potential till hur konventionella och hållbara produktattribut tillsammans med detaljerad information kan sammanföras till positiva köpintentioner i digitala sammanhang. Originalitet / Värde: Tidigare forskning utreder konsumenters inställning till hållbara produkter, gapet mellan hållbart intresse och handling samt hur konsumenter behandlar produktinformation. Mängden tidigare forskning som visar dock vilka hållbara produktattribut som kan sammanföras med produktattribut som tillfredsställer egennyttan hos konsumenter är sparsam. Det kan därför anses finnas ett behov för denna studie, då den ger företag insikter om hur kundfokus kan inkluderas i den hållbara utvecklingen. Särskilt relevant blir den för företag som riktat sig mot digitala försäljningskanaler och ämnar uppmuntra till hållbar konsumtion. / Purpose: This study aims to understand what sustainable product attributes that can be combined with attributes that fulfill consumer's self-interest. This is done by an examination of which product information about sustainable product attributes Swedish fashion consumers demand in a digital context to be able to make environmentally conscious decisions. Furthermore, which sustainable product information that can contribute with positive purchasing intentions. The purpose of this study is to be able to get an insight in how to reduce the distance between interest and action in terms of sustainable consumption.  Method: The method used for the study was inductive and the data was collected through a mixed method. Search of literature was followed by a quantitative survey, which was supplemented with qualitative semi-structured interviews. Participants were Swedish fashion consumers in the age range of 18-35. Quantitative and qualitative data were then compared and analysed against theories concluding that detailed product information can function as a substitute to physical interaction and that consumers do not prioritize sustainable product attributes which do not meet self-interest. The results were discussed and could then be presented as conclusions.  Results: The results show that product attributes which can be considered to satisfy self-interest are prioritised before sustainable product attributes, but sustainable information regarding fashion products might still increase buying intentions. Consumers experience, however, that companies do not communicate enough about sustainability and requests transparency and more detailed information. The results show a potential on how conventional and sustainable product attributes together with detailed information can be combined into positive buying intentions in a digital context. The study only focuses on fashion consumption on an online kontext. Neither did it evaluate the environmental impact of the different sustainable product attributes that were examined.  Originality / Value: Previous research investigates consumers’ attitudes towards sustainable products, the gap between interest and action as well as how consumers process information. There is a sparse amount of previous research that shows which sustainable product attributes consumers appreciate, which then can be combined with product attributes that satisfy the self-interest of consumers. There may therefore exist a need for this study, as it provides insights on how companies can include customer focus into their sustainable development. It will be particularly relevant for companies that have targeted digital sales channels and intend to encourage sustainable consumption.

This thesis investigates the impact management of product information has on the well-studied concepts of value propositions and business model innovation. Previously, it was not sufficiently documented how this type of information influences these concepts. As the pressure for digitalizing increases and traditional industries risk getting overthrown, understanding the factors affecting a company’s value proposition is more important than ever. To do this, this thesis outlines what product information is, for which stakeholders it matters, and during which phases in a product's lifecycle it has significance. Further, it gives four real-life examples of current use states (“as-is”) at a manufacturer of industrial motors - the case company. These examples are then analyzed and developed into proposed states (“to-be”) based on the findings from live observationat the factory, interviews with relevant stakeholders, and internal documents from the case company. Two main conclusions can be drawn based on the analysis performed on the real-lifeexamples. Firstly, it is found that well-managed product data have the ability to positively impact the value proposition. Secondly, it is found that it is too complex to determine which product information various stakeholders should have access to. It is more manageable to make all product data available to all stakeholders, via a data structure that minimizes the learning curve for users. The two conclusions have several practical and theoretical implications that are also discussed in detail. Finally, this thesis contributes to the field of industrial management by specifically outlining how product information affects value proposition, which in turn influences business models. / Denna rapport undersöker effekterna av hanteringen av produktinformation på de väletablerade koncepten värdeerbjudande och affärsmodellsinnovation. Tidigare har det inte varit tillräckligt dokumenterat hur denna typ av information påverkar dessa koncept. I takt med att trycket för digitalisering ökar och traditionella industrier riskerar kullkastas, så är förståelsen för de faktorer som påverkar ett företags värdeerbjudande viktigare än någonsin. För att göra detta har denna rapport undersökt vad produktinformation är, för vilka användare det har betydelse, och undervilka faser i en produktlivscykel som det har relevans. Dessutom ger rapporten fyra exempel, baserade från vekliga livet, på användarscenarion (“befintliga fall”) hos en tillverkare av industrimotorer - fallföretaget. Dessa exempel är sedan analyserade och utvecklade till föreslagna scenarion (“blivande fall”) baserade på de iakttagelser som gjorts på fallföretagets fabrik, intervjuer med relevanta parter och intern dokumentation från fallföretaget. Två huvudsakliga slutsatser kan konstateras baserat på analysen av dessa autentiska exempel. För det första konstateras att väl hanterad produktinformation kan påverka värdeerbjudandet positivt. För det andra upptäcks det att det är för komplicerat att avgöra vilken produktinformation olika parter bör ha tillgång till. I stället är det mer hanterbart att göra all produktinformation tillgänglig för allaparter, via en datastruktur som minimerar inlärningskurvan för användarna. Dessa två slutsatser har flertalet praktiska och teoretiska konsekvenser, som också diskuteras i detalj i rapporten. Avslutningsvis så bidrar denna rapport till fältet inom industriell ekonomi genom att särskilt diskutera hur produktinformation kan påverka värderbjudande, som i sin tur kan influera affärsmodeller.

Eftersom e-handeln ökar och returgraden likaså har returer blivit ett av e-handelns största miljöproblem, särskilt inom modebranschen. Detta arbete syftar därför till att undersöka vad som påverkar den höga returgraden samt hur olika medietekniska lösningar såsom produkt-information, kundrecensioner, storleksguider och virtuella provrum kan minska returgraden. I två fokusgrupper diskuterades arbetets syfte och resultatet sammanställdes sedan i en innehålls-analys. Med utgångspunkt i tidigare forskning och det empiriska materialet fördes en diskussion kring returorsaker samt medietekniska lösningar och deras potential. Både empiri och tidigare forskning vittnar om att storlek- och passformsproblem är de främsta orsakerna till retur och att medietekniska lösningar kan reducera antalet returer inom modebranschen genom att minska dessa problem. Dock behöver e-handelsföretag arbeta för att sådana lösningar ska användas i större utsträckning. / As the e-commerce is increasing so are the returns and returns have therefore become one of e-commerce biggest environmental problems, especially in the fashion industry. This study aims to investigate what influence the high return rate and how different technology solutions such as product information, customer reviews, size guides and virtual fitting rooms can decrease the return rate. In two focus groups the purpose of the study was discussed, and the result was compiled in a content analysis. Based on previous research and the empirical material a discussion was held about the causes of return as well as the media technology solutions and their potential. Both empirical material and previous research demonstrate size and fit problems as the main reasons for return and that media technology solutions can reduce the number of returns in the fashion industry. However, e-commerce companies must work for such solutions to be used to a greater extent.

Master’s thesis is focused on marketing research. The goal of this research is to know and describe the prospects and their perception of the product. Beside this goal the thesis demonstrates possible use of general statistical induction in the field of marketing research. The object of the research are social units carrying business in agriculture, forestry, timbering and building industry.

No / Marketers desire to utilise electronic word of mouth (eWOM) marketing on social media sites. However, not all online content generated by marketers has the same effect on consumers; some of them are effective while others are not. This paper aims to examine different characteristics of marketer-generated content (MGC) that of which one lead users to eWOM. Twitter was chosen as one of the leading social media sites and a content analysis approach was employed to identify the common characteristics of retweeted and favourited tweets. 2,780 tweets from six companies (Booking, Hostelworld, Hotels, Lastminute, Laterooms and Priceline) operating in the tourism sector are analysed. Results indicate that the posts which contain pictures, hyperlinks, product or service information, direct answers to customers and brand centrality are more likely to be retweeted and favourited by users. The findings present the main eWOM drivers for MGC in social media.

<p>I traditionella butiker kan kunden välja att ta kontakt med en försäljare för att få information om produkter eller tjänster. Försäljaren kan tipsa om lösningar efter kundens behov, även visa upp hur produkten fungerar och kunden kan själv välja vilken fråga han eller hon vill ha svar på. Den här mänskliga realtidskontakten kan i många fall vara värdefull och förtroendegivande för kunden. Vi antar att mervärden skapas för kunden när denne kan mötas ansikte mot ansikte och samtala med försäljare i en webbutik. För att se vilka dessa mervärden kunde vara, designade vi en prototyp för videosamtal. Denna implementerades i en fiktiv webbutik, som tillhandhöll kläder. Fem testpersoner fick agera presumtiva kunder och föra videosamtal med en professionell försäljare. Testpersonerna och försäljaren satt i olika rum och videosamtalen fördes via ett lokalt nätverk. Efter varje test genomfördes en kvalitativ intervju med testpersonerna och resultaten visade vilka mervärden som hade skapats. Ett mervärde var den personliga servicen med omedelbar feedback som försäljaren gav, i huvudsak via ljudkanalen. Videosamtalen medförde att kunderna kunde förhandla om priser och leveranstider. Ett annat var att kunder kunde få sina klädmått bedömda, men endast korrekt bedömda i två fall av fem. Ett annat visuellt mervärde var att kunden kunde be försäljaren att visa upp olika kombinationer av klädesplagg för att se hur de passade ihop med varandra. En testperson tyckte ett mervärde var att klädesplaggen kunde ses ur den vinkel som önskades. Ytterligare ett mervärde var att webbutiken upplevdes som trovärdig när försäljaren fanns i den som en fysisk gestalt. Testpersonerna tyckte att videosamtalen gjorde att de fick förtroende för webbutiken.</p> / <p>In retail shops in real life the customer can chose to contact a salesman to get information about products or services. The salesman can recommend solutions depending on the customers needs. The salesman can show the product to the customer and tell how it works. The customer on the other hand can ask questions to the salesman about the product to get the specific information that interests him or her. This human contact in real time can be valuable and confidece inspiring for the customer in many situations. We assume there are surplus values for the customer when he or she will meet face to face with a salesman via a video call in a webshop. To find out which these surplus values could be we designed a prototype for video calls. This was implemented in a fictitious webshop that supplied clothes. Five persons with experience from webbshopping acted customers and had video calls with a professional salesman. The customers and the salesman were in different rooms and could only communicate through a video call. Their computers were connected via a local network. We did qualitative interviews with every customer after each test and the results showed the surplus values that were generated. One surplus value was the personal service with instant feedback that the customer got from the salesman, mainly through the sound channel. The video calls resulted in negotiations about prices and delivery times. Another surplus value was that the salesman estimated the customers sizes in trousers and jackets. He only succeded in two cases of five. Another visual surplus value was that the customer asked the salesman to show articles of clothing in different combinations to see how them fit together. One person thought one surplus value was that he could see an article of clothing from whichever angle he wished. Further surplus value was that the customers experienced the webshop as trustworthy when the salesman was there as a physical human being. The customers thought that the video calls caused that they got confidence in the webshop.</p>

I traditionella butiker kan kunden välja att ta kontakt med en försäljare för att få information om produkter eller tjänster. Försäljaren kan tipsa om lösningar efter kundens behov, även visa upp hur produkten fungerar och kunden kan själv välja vilken fråga han eller hon vill ha svar på. Den här mänskliga realtidskontakten kan i många fall vara värdefull och förtroendegivande för kunden. Vi antar att mervärden skapas för kunden när denne kan mötas ansikte mot ansikte och samtala med försäljare i en webbutik. För att se vilka dessa mervärden kunde vara, designade vi en prototyp för videosamtal. Denna implementerades i en fiktiv webbutik, som tillhandhöll kläder. Fem testpersoner fick agera presumtiva kunder och föra videosamtal med en professionell försäljare. Testpersonerna och försäljaren satt i olika rum och videosamtalen fördes via ett lokalt nätverk. Efter varje test genomfördes en kvalitativ intervju med testpersonerna och resultaten visade vilka mervärden som hade skapats. Ett mervärde var den personliga servicen med omedelbar feedback som försäljaren gav, i huvudsak via ljudkanalen. Videosamtalen medförde att kunderna kunde förhandla om priser och leveranstider. Ett annat var att kunder kunde få sina klädmått bedömda, men endast korrekt bedömda i två fall av fem. Ett annat visuellt mervärde var att kunden kunde be försäljaren att visa upp olika kombinationer av klädesplagg för att se hur de passade ihop med varandra. En testperson tyckte ett mervärde var att klädesplaggen kunde ses ur den vinkel som önskades. Ytterligare ett mervärde var att webbutiken upplevdes som trovärdig när försäljaren fanns i den som en fysisk gestalt. Testpersonerna tyckte att videosamtalen gjorde att de fick förtroende för webbutiken. / In retail shops in real life the customer can chose to contact a salesman to get information about products or services. The salesman can recommend solutions depending on the customers needs. The salesman can show the product to the customer and tell how it works. The customer on the other hand can ask questions to the salesman about the product to get the specific information that interests him or her. This human contact in real time can be valuable and confidece inspiring for the customer in many situations. We assume there are surplus values for the customer when he or she will meet face to face with a salesman via a video call in a webshop. To find out which these surplus values could be we designed a prototype for video calls. This was implemented in a fictitious webshop that supplied clothes. Five persons with experience from webbshopping acted customers and had video calls with a professional salesman. The customers and the salesman were in different rooms and could only communicate through a video call. Their computers were connected via a local network. We did qualitative interviews with every customer after each test and the results showed the surplus values that were generated. One surplus value was the personal service with instant feedback that the customer got from the salesman, mainly through the sound channel. The video calls resulted in negotiations about prices and delivery times. Another surplus value was that the salesman estimated the customers sizes in trousers and jackets. He only succeded in two cases of five. Another visual surplus value was that the customer asked the salesman to show articles of clothing in different combinations to see how them fit together. One person thought one surplus value was that he could see an article of clothing from whichever angle he wished. Further surplus value was that the customers experienced the webshop as trustworthy when the salesman was there as a physical human being. The customers thought that the video calls caused that they got confidence in the webshop.

The exceeding supply over demand and very hard competitive conditions are nowadays the main features of the majority of sectors. A successful company is the one that is able to satisfy specific customers' needs, the one that has efficient cooperation with its suppliers throughout the whole supply chain and also the one that is able to speed up the in-house information exchange. Thus the company has to seek constantly new and innovative solutions. This is not possible without standardization and automatization of business processes. This master's thesis is dedicated to one of the possible solutions -- the Product Information Management (PIM). Since it is intended for business managers (without deep IT knowledge), at the beginning it answers the question why it is so important to know master data and to manage it. It specializes in managing product data, brings its comprehensive overview and identifies the advantages and drawbacks of the implementation as well as financial and organizational impacts. The consecutive chapter deals with simplified yet applicable approach to data management analysis (with emphasis on the PIM) and based on research, it mentions main mistakes of the implementation. In addition to the overview of main vendors of the PIM solution, it presents the latest trends in the PIM. Besides internal data synchronization, the thesis analyses several product standards -- the fundamental step towards external data synchronization, the key topic of the practical part. The whole thesis is conceived to provide an organization with a simple yet compact and therefore very effective tool for master product data insight and thus to help it to gain a competitive advantage.